Why Are Electricity Prices Increasing? An Industry-Wide Perspective Prepared by: Prepared for:

Why Are Electricity Prices Increasing? An Industry-Wide Perspective Prepared by: Prepared for:
Why Are Electricity Prices Increasing?
An Industry-Wide Perspective
Prepared by:
Gregory Basheda
Marc W. Chupka
Peter Fox-Penner
Johannes P. Pfeifenberger
Adam Schumacher
The Brattle Group
Prepared for:
JUNE 2006
The Edison Foundation is a nonprofit organization dedicated
to bringing the benefits of electricity to families, businesses,
and industries worldwide. Furthering Thomas Alva Edison’s spirit of invention, the
Foundation works to encourage a greater understanding of
the production, delivery, and use of electric power to foster
economic progress; to ensure a safe and clean environment;
and to improve the quality of life for all people.
The Edison Foundation provides knowledge, insight,
and leadership to achieve its goals through research,
conferences, grants, and other outreach activities.
The Brattle Group
The Brattle Group provides consulting services and
expert testimony in economics, finance, and regulation to
corporations, law firms, and public agencies worldwide.
Our principals are internationally recognized experts, and
we have strong partnerships with leading academics and
highly credentialed industry specialists around the world.
The Brattle Group has offices in Cambridge, Massachusetts;
San Francisco; Washington, D.C.; Brussels; and London.
Detailed information about The Brattle Group is available
at www.brattle.com.
© 2006 by The Edison Foundation.
All Rights Reserved under U.S. and foreign law, treaties and conventions. This Work cannot be reproduced, downloaded,
disseminated, published, or transferred in any form or by any means without the prior written permission of the copyright owner
or pursuant to the License below.
License – The Edison Foundation grants users a revocable, non-exclusive, limited license to use this copyrighted material for
educational and/or non-commercial purposes conditioned upon the Edison Foundation being given appropriate attribution for
each use by placing the following language in a conspicuous place, “Reprinted with the permission of The Edison Foundation.”
This limited license does not include any resale or commercial use.
Published by:
The Edison Foundation
701 Pennsylvania Avenue, N.W.
Washington, D.C. 20004-2696
Phone: 202-347-5878
Table of Contents
Chapter 1: Introduction...........................................................................................................................................1
Introduction and Purpose ........................................................................................................................................1
Overview of Findings .............................................................................................................................................2
Electricity Remains An Excellent Value ................................................................................................................5
The Structure of This Report ..................................................................................................................................6
Chapter 2: Increased Fuel Prices Drive Utility Costs ...........................................................................................9
Utilities’ Rising Costs Are Primarily Due to Higher Fuel Prices ...........................................................................9
Power Generation Fuel Costs ...............................................................................................................................10
Natural Gas ...........................................................................................................................................................11
Oil .........................................................................................................................................................................13
Coal.......................................................................................................................................................................15
Nuclear Fuel..........................................................................................................................................................19
Purchased Power Costs.........................................................................................................................................20
Wholesale Prices Are Increasing and Becoming More Volatile...........................................................................21
Chapter 3: Drivers of Electricity Demand ...........................................................................................................25
Increasing Demand for Power ..............................................................................................................................25
The Effect of Price Increases on Power Demand .................................................................................................30
The Impact of Demand-Reduction Programs .......................................................................................................31
Historical Energy and Demand Savings from DSM Programs.............................................................................33
Potential Energy Savings from DSM Programs ...................................................................................................33
Savings from Appliance and Equipment Standards..............................................................................................34
EPA ENERGY STAR® Program..........................................................................................................................36
Demand-Response Programs ................................................................................................................................36
Real-Time Pricing.................................................................................................................................................39
Conclusion ............................................................................................................................................................40
Chapter 4: Generation Investment .......................................................................................................................41
Generation Additions: Past, Present, and Future .................................................................................................41
Coal-Fired Generation ..........................................................................................................................................44
Nuclear Power Plants............................................................................................................................................45
Renewables ...........................................................................................................................................................46
Renewable Energy Standards ...............................................................................................................................46
Green Electricity Marketing .................................................................................................................................48
On-Site Customer Generation...............................................................................................................................48
Chapter 5: Transmission Investment ...................................................................................................................51
Overview of the Transmission Grid......................................................................................................................51
Transmission Investment Trends and Drivers ......................................................................................................52
Transmission Investment Looking Forward .........................................................................................................54
Factors Driving Increased Transmission Investment............................................................................................55
iii
Table of Contents
Policy Initiatives to Facilitate Transmission Investment ......................................................................................57
Transmission Grid and Retail Rates .....................................................................................................................57
Transmission Grid of the Future ...........................................................................................................................58
Chapter 6: Distribution Investment......................................................................................................................63
Trends in Distribution System Investment ...........................................................................................................63
Need to Modernize Distribution Systems .............................................................................................................64
Investments in Metering .......................................................................................................................................66
Minimizing Outage Costs .....................................................................................................................................67
Chapter 7: Environmental Investments ...............................................................................................................69
Overview...............................................................................................................................................................69
Utility Environmental Protection Investments and Results ..................................................................................70
Environmental Costs and Rate Impacts ................................................................................................................72
Climate Change and Electric Generation..............................................................................................................75
New Generating Technologies..............................................................................................................................77
Costs of CO2 Controls...........................................................................................................................................78
Chapter 8: Financial Condition and Outlook ......................................................................................................79
The Industry’s Financial Condition During the Last Decade ...............................................................................79
Utility Credit Ratings ..................................................................................................................................79
Earned and Allowed Returns on Equity ......................................................................................................81
Increasing Risks ..........................................................................................................................................82
Operating Cash Flows and Capital Spending ..............................................................................................83
Summary: Utilities’ Financial Condition over the Past 10 Years................................................................85
Financial Outlook: The Challenges Ahead ...........................................................................................................86
The Outlook for Utility Credit Ratings and Earned Returns ................................................................................87
Increasing Financing Costs..........................................................................................................................89
Chapter 9: Cost Recovery, Investment, and Rates in Perspective .....................................................................93
Historical Prices in Perspective ............................................................................................................................93
Electricity Prices by Customer Class....................................................................................................................93
How Electricity Prices Increase ............................................................................................................................96
The Role of Rate Increases ...................................................................................................................................97
The Long-Term Benefits of Appropriate Rate Treatment ....................................................................................97
Appendix A: Household Power Use: Past, Present, and Future ........................................................................99
Appendix B: Impacts of Price Increases on Electricity Demand Growth Forecasts ......................................103
Appendix C: Discussion of Historical Transmission Investment Trends ........................................................107
iv
CHAPTER 1
Introduction
Introduction and Purpose
For more than a century, the electric power industry has supplied the United States with abundant and
reliable electricity. The industry that brought “smokeless light” to American cities in the late 1800s now
supplies the power for more than 176 million personal computers and a national network of 208 million
cellular phones, contributing to both industrial productivity and consumer comforts that enhance our
standard of living.
The power industry now faces an unprecedented challenge. At a time of record high fuel prices, historic
environmental challenges, and industry structural change, the nation’s demand for reliable electric power
continues to grow. While much of the nation’s power infrastructure is aging, the industry must keep up with
the need for more capacity, increased reliability and power quality, and lower environmental impacts. Thus,
the industry must invest in a new generation of power plants, environmental controls, transmission lines, and
distribution system expansions and upgrades.
While these new investments will maintain reliability, diversify our fuel mix, and increase environmental
performance, they come with added costs. Electricity price increases are occurring across the United States,
among all types of electricity providers, to one degree or another. The extent to which increasing utility
costs are recovered in rates will determine the financial condition of the industry and affect its ability to
make future generation, transmission, distribution, and environmental investments in a timely manner.1
With appropriate rate treatment, the industry will continue to provide reliable services at reasonable costs.
Conversely, if segments of the industry become unable to finance new investments in a timely or costeffective manner, the ultimate costs will be borne by the local economies and consumers served by these
utilities, as well as by utility shareholders. Failure to receive adequate rate treatment could impact the
quality of service, impair the ability of the utility industry to meet growing demands for clean, reliable
power, and undermine the financial health of the utility industry.
This report examines the factors underlying the recent increases in electricity prices and the potential impacts
of these factors on the industry’s financial condition. We focus primarily on cost changes experienced over
the past five years and the projected trends in these costs over the next decade. The trends we examine affect
all electricity suppliers, while the focus of this paper is the impact of higher costs and capital expenditures on
1
Throughout this report, electricity rates will refer to the retail price of electric service provided by utilities subject to costbased regulation, including utilities with residual, regulated services in restructured states. The term electricity prices is
broader, and includes both regulated rates and retail prices charged by electricity suppliers not subject to cost-of-service
ratemaking.
1
Chapter 1: Introduction
rates that require regulatory approval. Our analysis examines the investor-owned segment of the industry as
a whole, using a national perspective. While the circumstances of each provider’s costs and prices are
unique, and must be considered individually, several common factors and trends are influencing the entire
industry. Nevertheless, the analyses and conclusions in this report should not be construed as applying to any
particular utility without further careful consideration.
Overview of Findings
Fuel and Purchased Power Cost Increases Have Been Enormous and Are the Largest Cause of Recent
Electric Cost Increases. On an industry-wide basis, our analysis finds that fuel and purchased power costs
account for roughly 95 percent of the cost increases experienced by utilities in the last five years. The
increases in the cost of these fuels have been unprecedented by historical standards, affecting every major
electric industry fuel source:
ƒ Natural gas, which accounts for nearly 20 percent of all generation, experienced a more than 100percent increase in spot prices between 2003 and 2005 and a more than 300-percent increase since
1999. Real natural gas prices are now at their highest level in modern history. High and volatile gas
prices have a particularly strong impact on electricity prices because gas-fired generators set the prices
for a large percentage of the time in many short-term or spot power markets around the country.
ƒ Oil, which is still a significant utility fuel in several parts of the country, is now at record price levels.
The prices of oil-based fuels delivered to electric generators rose about 50 percent between 2003 and
2005, and are now at the highest nominal levels ever recorded. Increased oil prices also have a
significant impact on other fuel costs; for example, they drive up the costs of mining and shipping
coal.
ƒ Coal, which accounts for half of all power produced in the United States today, has risen 20 percent in
delivered price in the last two years alone. In some areas, the increase has been much higher. For
example, spot coal prices from the Powder River Basin have increased about 100 percent since 2003.
ƒ The price of uranium, the primary component of nuclear fuel, which represents 19 percent of all
generation, also has increased by about 40 percent since 2001.
These fuel price increases, in turn, have impacted the cost of power purchased by many utilities. The price
of purchased spot power has increased between 200 and 300 percent in many power markets across the
United States. Finally, the industry is using increasing amounts of renewable and distributed generation
resources, which have valuable attributes but generally cost more than conventional energy sources.
Additional Generating Plants Will Be Needed To Meet Demand. The Energy Information Administration
(EIA) and the North American Electric Reliability Council (NERC) both project that more than 50,000
megawatts (MW) of new power plants will be needed to meet demand growth through the year 2014. There
are several aspects of the next wave of generation investments worthy of note:
ƒ Prompted by recent natural gas prices and prospects for continued demand growth, new baseload coal
plants are being proposed and/or built for the first time in more than a decade. More than a quartercentury after the last nuclear plant was ordered, new nuclear plants are under active consideration.
2
Why Are Electricity Prices Increasing? An Industry-Wide Perspective
The Energy Policy Act of 2005 (EPAct 2005), in conjunction with other federal programs, will help
reduce the costs and risks of building these generating additions, which are larger, are more capitalintensive, and have a longer lead time than the natural gas-fired units the industry built over the past
decade.
ƒ New generation investment varies substantially by region and by each utility’s present fuel mix. Some
areas of the country remain chronically short on power and will need a variety of new resources to
meet demand. Other regions are now strongly reliant on gas-fired generation, and may add coal-fired
capacity to diversify the fuel mix and reduce the total cost of electricity. Finally, nearly half of the
states now require utilities to build or purchase energy from renewable electric generators, which will
help diversify their fuel mix but add to overall costs.
ƒ Uncertainties over future fuel prices, climate change policies, technological progress in all the major
power technologies, and the impact of higher prices on power demand create substantial risks
enveloping new generation investments. These risks add to the cost of financing these investments.
ƒ The need for additional generation and transmission capacity will be mitigated by demand and energy
reductions achieved through the price elasticity impact of rising prices and through a variety of
conservation, energy efficiency, and demand-response programs. However, there still will be a need
in the future for utilities to make major investments in generation and transmission capacity.
Increased Transmission Investments Are Necessary. After a long period of decline, transmission
investment began a significant upward trend in the year 2000, totaling nearly $18 billion in the period 1999
to 2003. A recent Edison Electric Institute (EEI) survey shows that its members have spent and plan to
spend nearly $29 billion on transmission over the period 2004 to 2008, a 60-percent increase over the
previous five years. NERC projects that almost 12,500 miles of new transmission will be added by 2014, an
increase of 5.9 percent of total U.S. circuit miles of high-voltage [230 kilovolts (kV) and above] transmission
lines.
ƒ These increased investments are prompted in part by the larger scale of the next wave of baseload
generation additions and the fact that these additions are occurring farther from load centers. This is
creating transmission projects that are larger and more costly than the average project over the past 20
years.
ƒ New government policies and industry structures also will contribute to greater transmission
investment. EPAct 2005 creates new incentives and siting processes that facilitate and promote
transmission investment. In many parts of the country, transmission planning has been formally
regionalized, and power markets create greater price transparency that highlights the value of
transmission expansion in some instances.
Sales Growth, the Demand For Higher Quality Power, and Storm Recovery Costs Are Driving
Distribution Investment. Industry spending on the distribution systems that deliver power to each customer
has followed a generally steady upward trend for the past 20 years. Between 2000 and 2004, distribution
investment increased from about $10.5 billion to $12.5 billion, a 19-percent increase.
3
Chapter 1: Introduction
ƒ Many of these investments are in new technologies that increase the quality of delivered power to
ubiquitous digital circuits. Other investments are being made to make the distribution system more
automated, information-rich, and responsive to outages and customer needs. For example, some
automated distribution systems provide customers with the ability to monitor and control their energy
usage on specific processes and appliances, depending on real-time prices and other factors.
ƒ Additional large distribution system expenditures have been necessitated by widespread hurricane and
storm damage experienced in the southeastern United States during 2004 and 2005, which impacted
energy and materials costs across the nation.
Environmental Investments Add Significant Costs. New environmental requirements, including recently
finalized federal rules and state-level requirements that often are more stringent and less flexible, are
prompting substantial environmental investments. These investments include more than $43 billion in
planned capital costs for emissions reduction technologies from 2005 to 2018, primarily retrofit equipment to
further control air emissions from existing coal-fired power plants. These investments, while large, could be
dwarfed by the costs of complying with potential mandatory carbon dioxide (CO2) emission reductions, as
such policies have recently been proposed and considered in Congress.
The Utility Industry’s Overall Financial Condition Is Sound, Though Not As Secure As It Had Been
Before Prior Periods of Capital Investment. With reasonable cost recovery, the industry as a whole should
have the ability to make the necessary, cost-effective investments. However, the industry has
proportionately less “headroom” to make investments without rate relief, and certain portions of the industry
are already below investment grade and therefore cannot weather greater financial impairment.
ƒ The fraction of utilities rated BBB+ or above by Standard and Poor’s, which was 75 percent prior to
the 1990s, is now only about 40 percent. As of 2005, nearly 20 percent of all utilities were below
investment grade. The credit ratings of independent power producers are significantly worse.
ƒ Between 1999 and 2005, interest rates, allowed utility returns on equity (ROEs), and earned ROEs all
trended downward at similar rates that enabled earned ROEs to remain reasonably close to allowed
ROEs. However, the future prospects for earnings, absent adequate rate increases, are worse. Costs
are rising much faster than revenues, and interest rates are no longer on a downward trend.
ƒ The reduced financial stability of the industry is reflected in the “beta” of utility stocks—a measure of
the proportionate riskiness of these stocks compared to the overall market. Value Line’s estimate of
the average industry beta has increased from 0.67 in 1995 to 0.87 in 2005, an increase of nearly 30
percent in a decade.
ƒ The operating cash flows of utilities in 2005 were insufficient to cover their capital expenditures and
higher operating costs. Utility cash flows were about $10 billion less than the sum of operating and
capital costs in 2005, and this gap could widen significantly during the next several years as regulated
utilities undertake expenditures for infrastructure development and environmental improvements.
The overall picture emerging from these conclusions is that the electric power industry faces a situation in
which significant investments are needed, and rate increases will be necessary to finance them. These
investments will diversify supply away from natural gas, reduce future fuel costs, provide greater reliability
4
Why Are Electricity Prices Increasing? An Industry-Wide Perspective
and power quality, and lessen environmental impacts. Without these investments, one or more of these
investment objectives will be impaired.
Electricity Remains An Excellent Value
Even with price increases, electric power continues to grow in value to American consumers and the
American economy. Since 1940, the percentage of U.S. energy consumed in electric form has quadrupled.
Electricity demand growth tracks Gross Domestic Product (GDP) growth much more closely than any other
source of energy, highlighting its role as a key driver of economic growth and productivity.
As electricity use has grown in economic value, its inflation-adjusted cost has been declining. From 1985 to
2000, average electricity prices rose 1.1 percent per year, less than half the average inflation rate of 2.4
percent. Figure 1-1 shows real electricity prices (in year 2005 dollars) by customer class over the period
1960 through 2005. After peaking in the early 1980s, average real prices had fallen by about 25 percent by
2005. And, compared with prices of other consumer goods and services, electricity prices have risen more
slowly. This is shown in Figure 1-2, which uses 1970 as a base year for price indices for electricity,
gasoline, natural gas, and medical care. Finally, despite increased household electricity consumption,
electricity bills have become a smaller fraction of household budgets. American homes use 21 percent more
electricity today than they did in 1978. Yet even with 21 percent greater use, the portion of our household
budget that we devote to our power bill has declined, from 3.7 percent to 3.0 percent over the same period.
Figure 1-1
U.S. Electricity Prices by Class of Customer (Real 2005 Dollars)
14
Average Price (¢/kWh)
U.S. Average
Commercial Price
12
U.S. Average
Residential Price
10
8
U.S. Average
Industrial Price
U.S. Average
Total Price
6
2005
2002
1999
1996
1993
1990
1987
1984
1981
1978
1975
1972
1969
1966
1963
1960
4
Sources: EIA Annual Energy Review 2004, EIA Monthly Energy Review March 2006, and U.S. Bureau of Labor Statistics.
Note: Real dollars calculated from U.S. GDP deflator.
5
Chapter 1: Introduction
Figure 1-2
Comparison of Electricity and Other Consumer Price Trends
(1970 to 2005)
1,100
Natural
Gas
1,000
Base 1970=100
900
Medical
Care
800
700
600
Gasoline
500
400
300
Electricity
All Items
200
2004
2002
2000
1998
1996
1994
1992
1990
1988
1986
1984
1982
1980
1978
1976
1974
1972
1970
100
Sources: EIA Annual Energy Review 2004, EIA Monthly Energy Review March 2006, and U.S. Bureau of Labor Statistics.
Americans already own an ever-growing array of devices that provide services unimagined even a few years
ago, from multi-function cell phones to MP3 players. Future American homes will contain intelligence and
sensors that will manage and reduce energy costs substantially. This will include products such as advanced
meters and “smart” appliances that interact seamlessly with the power grid and service providers.
The next power investment wave will also provide American businesses with more options and greater
productivity. Digital-quality power now represents 10 percent of total electrical load in the United States and
is expected to reach 30 percent by 2020.2 At the same time, underinvestment in transmission and distribution
is estimated to cost the American economy at least $20 billion a year—a figure certain to grow if
transmission and distribution infrastructure investment does not keep pace with demand.
The Structure of This Report
In this report, we examine the causes and potential effects of electricity price increases. We begin in Chapter
2 by examining recent trends and projected changes of the two core components of most utilities’ operating
costs: fuel and purchased power. Specifically, the recent increases in the price of utility fuels—natural gas,
oil, coal, and nuclear fuel—are highlighted and explained. These increased fuel costs drive similar increases
in the cost of power purchased in wholesale markets.
2
U.S. Department of Energy, Office of Electric Transmission and Distribution, “Grid 2030” – A National Vision For
Electricity’s Second 100 Years, July 2003, p.3.
6
Why Are Electricity Prices Increasing? An Industry-Wide Perspective
In Chapter 3, we focus on increasing demands for reliable electric power, based upon the long-term
relationship between economic growth, technological progress, and the increased electrification of the
economy. A series of demand projections are presented, and we discuss the impact of higher electricity
prices and demand-reduction and load management programs on expected demand growth.
Next, we consider the need for infrastructure investment by electric utilities. In Chapter 4, we look at
generation-baseload investment, advancements in renewables, and on-site customer generation to assist in
capacity needs. In Chapter 5, we examine transmission—the need for investment based on recent trends and
the need to enhance wholesale market operation. In Chapter 6, we look at distribution investments and the
need for better power delivery. In Chapter 7, we examine the costs incurred by utilities as they meet new
environmental requirements.
In Chapter 8, we look at the financial condition of utilities and how that condition impacts the ability of
utilities to pursue investment. We review the trends in utility credit ratings, the earned and allowed returns
on equity, and the increasing financial risks of utilities.
In Chapter 9, we conclude the report by putting cost recovery and electric rates in perspective, and highlight
the long-term benefits of making necessary investments in generation, transmission, distribution, and
environmental technologies.
7
CHAPTER 2
Increased Fuel Prices
Drive Utility Costs
Utilities’ Rising Costs Are Primarily Due to Higher Fuel Prices
Between 2002 and 2005, annual operations and maintenance (O&M) expenses for investor-owned utilities
(IOUs) increased approximately 22 percent. This section analyzes the core reasons for this increase. It
begins by illustrating the primary importance of fuel and purchased power on overall expense trends. Next, a
review of fuel price trends begins to explain the higher expenses facing utilities. Finally, a review of
wholesale power markets provides context for rising purchased power expenses. The wide prevalence of
fuel and purchased power adjustment clauses that serve to reflect these input costs in electric rates has
greatly influenced much of the rate increases that already have occurred.
Increases in fuel and purchased power costs account for virtually the entire rise in operating expenses for
electric utilities. Figure 2-1 illustrates FERC Form 1 data compiled for a sample of more than 180 utilities
serving retail load. By 2005, fuel and purchased power expenses amounted to 71 percent of total O&M
expenses, compared to 66 percent of total O&M expenses in 2002. Fuel and purchased power expense
growth essentially explains all of the 22-percent increase in utilities’ expenses from 2002 to 2005. While
transmission expenses increased at a slightly higher rate than fuel and purchased power, most likely due to
dramatic upturns in transmission investment observed between 2000 and 2005, this category still only
represented about four percent of total operating expenses in 2005. Distribution expenses remained
essentially flat, and other expenses actually declined two percent over this time period.
The sharp rise in utilities’ fuel costs impacts utilities and customers in different ways in various regions. For
states that have not pursued retail restructuring, fuel expenses for utility-owned generation constitute a core
component of expenses, often passed through to consumers in fuel adjustment clauses (FACs). In states that
have pursued restructuring, many utilities face higher purchased power expenses from wholesale markets,
which comprise a major portion of their supply. Whatever the mechanism through which utilities face rising
fuel and purchased power costs, the stakes are extremely high. In analyzing rising unit costs, a major credit
rating agency stated: “[T]he ramifications of higher gas commodity prices and the related effects on the
prices of coal, emission credits and wholesale electric power are tipping the balance toward greater risk for
regulated gas and electric utilities and for those generators most dependent on natural gas.”3
3
Fitch Ratings, “Rising Unit Costs: A Threat to Utility Sector Credit,” November 4, 2005, p. 1.
9
Chapter 2: Increased Fuel Prices Drive Utility Costs
Figure 2-1
Drivers of Electric Utility Operations and Maintenance Expenses
160
95% of $26 billion increase in electric O&M expenses is
driven by increases in fuel and purchased power expenses
140
$ Billions
120
100
104
87
92
30
32
32
30
7
4
8
4
8
5
7
5
2002
2003
2004
2005
79
80
60
40
20
0
Source: FERC Form 1.
Transmission
Distribution
Other
Fuel and Purchased Power
Power Generation Fuel Costs
The majority of electric generation capacity uses fossil or nuclear fuel to create heat for steam turbines, or
burns fossil fuel to drive combustion turbines. Figure 2-2 provides a breakdown of electric net generation by
fuel type in 2005. The combination of coal, natural gas, nuclear, and oil-fired generation accounts for more
than 90 percent of U.S. national net generation. Accordingly, the costs of these fuels will be the focus of this
section.
Figure 2-2
Net Generation by Energy Source 2005
Nuclear
19%
Coal
50%
Natural Gas
19%
Other Oil
2% 3%
Source: EIA (preliminary 2005 data).
10
Hydro
7%
Why Are Electricity Prices Increasing? An Industry-Wide Perspective
Natural Gas
While accounting for only 19 percent of U.S. electric net generation, natural gas exerts a disproportionate
influence on electricity prices in the wholesale market because it represents the incremental generation in
most high-demand hours. As the price of purchased wholesale power and retail real-time prices are
increasingly based on power spot markets, the marginal cost of the most expensive (“marginal”) unit that sets
the hourly price has a significant effect on both hourly and longer-term wholesale contract purchases.
The price of natural gas has always been somewhat higher and more volatile than coal, but over the past few
years natural gas price levels and volatility have increased dramatically. This has occurred during an era
when natural gas-fired capacity has dominated the new capacity market, leading to new plants running less
than expected or even idled in some cases.
Figures 2-3 and 2-4 depict movements in delivered prices of natural gas to electric generators and spot gas
prices, respectively. Natural gas prices delivered to generators reflected stable to declining price levels from
the late 1980s to the late 1990s. Coinciding with the surge in natural gas-fired combined-cycle capacity
brought on-line in the early 2000s, prices increased dramatically. Between 2003 and 2005, gas prices
delivered to electric generators increased more than 50 percent, while spot prices surged more than 100
percent.
Unlike coal, there are a variety of other end-use segments that consume natural gas. In 2004, the electric
power sector accounted for about 26 percent of natural gas delivered to end-use customers. For industrial
customers, the largest single sector of natural gas consumers, trends in overall economic growth dictate
demand, while weather drives demand for natural gas heating among residential customers. For example, the
aberrant spike in Henry Hub spot prices in 2003 reflected below-average temperatures during the winter
months. The combination of spikes related to weather, continued economic growth, and a dramatic
expansion of natural gas-fired combined-cycle generating capacity all contribute to pressures on the supply
of natural gas.4 In addition, critical infrastructure disturbances in the Gulf Coast due to the hurricanes of
2005 contributed to the significant volatility observed for spot prices in that year.
4
http://www.eia.doe.gov/pub/oil_gas/natural_gas/feature_articles/2006/ngmarkets/ngmarkets.pdf.
11
Chapter 2: Increased Fuel Prices Drive Utility Costs
Figure 2-3
Historical Delivered Natural Gas Prices ($ Nominal)
9.00
Gas Price - $ / MMBtu
8.00
7.00
6.00
5.00
4.00
3.00
2.00
1.00
2005
2003
2001
1999
1997
1995
1993
1991
1989
1987
1985
1983
1981
1979
1977
1975
1973
0.00
Year
Source: EIA Monthly Energy Review.
Figure 2-4
Historical Henry Hub Spot Prices
20.00
18.00
16.00
$/MMBtu
14.00
12.00
10.00
8.00
6.00
4.00
2.00
3/7/06
8/25/05
2/24/05
8/26/04
2/26/04
8/28/03
2/27/03
8/29/02
2/28/02
8/30/01
3/1/01
0.00
Sources and Notes:
Platts Gas Daily.
Excludes 8/30/05, 9/27/05-10/7/05 when Henry Hub operations were disrupted due to damage caused by Hurricanes Katrina and
Rita.
A great deal of uncertainty exists about the direction of natural gas prices delivered to electric generators.
Figure 2-5 depicts a variety of forecasts through 2015, and shows that EIA's Annual Energy Outlook 2006,
published in February 2006, predicted significant declines in prices from 2005 peaks. However, EIA's Short
Term Energy Outlook, published more frequently and updated in May 2006, shows prices dipping and then
returning to their high levels in 2007. As shown, EIA’s Annual Energy Outlook 2006 and Global Insight,
12
Why Are Electricity Prices Increasing? An Industry-Wide Perspective
Inc.5 predict steady declines in the price of natural gas, both in nominal and real terms. On the other hand,
Energy and Environmental Analysis, Inc.6 predicted an opposite picture with prices remaining at high levels.
Finally, the simple average of NYMEX monthly futures prices as of May 4, 2006, for Henry Hub is plotted
for the years 2007 to 2011. While not a measure of delivered prices, the NYMEX trend is downward but at
high absolute levels. Clearly, the significant volatility in natural gas prices and different views regarding
longer-term structural market issues, such as the amount and price of liquefied natural gas (LNG) imports,
contribute to a wider range of uncertainty for the future of natural gas prices.
Figure 2-5
Forecasts of Delivered Natural Gas Prices ($ Nominal)
11.00
$ / MMBtu
10.00
9.00
8.00
7.00
Energy and
Environmental
Analysis
Forecast
EIA Short-Term
Energy Outlook
Average of Monthly
NYMEX Futures (5/4/06)
Historical
Data
EIA Reference, High
EIA
High
andReferenec,
Low Economic
and
Low Economic
Growth
Cases
Growth Cases
6.00
Global Insight,
Inc. Forecast
2015
2014
2013
2012
2011
2010
2009
2008
2007
2006
2005
2004
2003
5.00
Year
Sources and Notes:
EIA Annual Energy Outlook 2006, unless noted.
Real prices converted to nominal using forecast GDP deflator.
Oil
While representing only three percent of net generation, oil-fired capacity is quite prevalent in some regions
and represents the marginal price-setting fuel in many peak hours. For example, oil-fired generation
produces nearly 14 percent of net generation in New York and 10 percent of net production in New
England.7 In 2005, oil-fired generation units were on the margin during 11 percent of the time in the
Pennsylvania, New Jersey, and Maryland (PJM) market, a large energy market in the mid-Atlantic and
Midwest region.8 In addition to their role in influencing peak electricity prices in several regions, oil price
increases also translate into mining and transportation cost increases that impact the delivered price of coal
and other utility costs.
5
6
7
8
Published in August 2005.
Published in October 2005.
Energy Information Administration, Annual Energy Outlook 2006, February 2006.
PJM, State of the Market Report 2005, p. 86.
13
Chapter 2: Increased Fuel Prices Drive Utility Costs
The prices of petroleum generation fuel (#6 residual and #2 distillate) in the electric generating sector have
mirrored the significant price increases in all petroleum products. Figures 2-6 and 2-7 present EIA’s
historical and projected prices for petroleum fuel delivered to electric generators. Historically, average
petroleum prices have tracked the movements of natural gas prices, with levels increasing nearly 50 percent
between 2003 and 2005. EIA's projections of delivered petroleum prices show steady to declining nominal
levels in the near term,9 with steadily increasing prices after 2010. Thus, according to EIA, regions that are
more exposed to oil-fired generation can expect elevated, and ultimately rising, petroleum fuel product prices
to influence wholesale electric market peak prices over the next five to 10 years.
Figure 2-6
Historical Delivered Petroleum Prices ($ Nominal)
7.00
6.00
$ / MMBtu
5.00
4.00
3.00
2.00
1.00
Source: EIA Monthly Energy Review.
Weighted average of residual and distillate petroleum fuel.
9
2005
2003
2001
1999
1997
1995
1993
1991
1989
1987
1985
1983
1981
1979
1977
1975
1973
0.00
Year
Once again, EIA’s most recent Short Term Energy Outlook (May 2006) forecasts higher delivered petroleum prices to
electric generators in 2006 and 2007.
14
Why Are Electricity Prices Increasing? An Industry-Wide Perspective
Figure 2-7
Forecasts of Delivered Petroleum Prices ($ Nominal)
9.00
EIA Reference, High
and Low Economic
Growth Cases
8.50
$ / MMBtu
8.00
7.50
7.00
6.50
Historical
Data
6.00
5.50
2015
2014
2013
2012
2011
2010
2009
2008
2007
2006
2005
2004
2003
5.00
Sources and Notes:
Year
EIA Annual Energy Outlook 2006, unless noted.
Real prices converted to nominal using forecast GDP deflator.
Coal
Half of all net generation in the United States is produced from coal, and consumption of coal by electric
utilities accounts for about 92 percent of total U.S. coal consumption.10 This section will analyze historical
trends in both spot prices for coal and delivered prices to electric generators, additional cost drivers
associated with emissions allowances prices, and recent forecasts for trends in both spot and delivered prices.
The vast majority of coal volumes are under long-term, multi-year contracts. For example, between 2000
and 2002, only 28 percent of coal purchases in Central Appalachia and 18 percent of purchases from the
Powder River Basin were made on the spot market.11 As a consequence, price increases in the spot market
(to the extent that they persist) will emerge as higher delivered prices over time, as long-term contracts
gradually expire and current market conditions influence new contract prices.
How have spot prices and delivered prices moved in recent years? Figures 2-8 and 2-9 depict movements in
delivered contract prices and coal spot prices to electric generators, respectively. Delivered coal prices,
which reflect contracts that bind the majority of coal deliveries, declined in nominal terms starting in 1985
for approximately 15 years. However, between 2003 and 2005, delivered coal prices to electric generators
increased by more than 20 percent.
10
Energy Information Administration, U.S. Coal Supply and Demand Review 2005,
http://www.eia.doe.gov/cneaf/coal/page/special/feature.html.
11
Energy Information Administration, U.S. Coal Prices, http://www.eia.doe.gov/cneaf/coal/page/uscoal.pdf#page=2.
15
Chapter 2: Increased Fuel Prices Drive Utility Costs
As expected, this movement in delivered average prices understates the run-up in the spot price of coal as
illustrated in Figure 2-9. This EIA chart illustrates that spot prices have risen in every major geographic
region of coal production. For example, prices at the Powder River Basin have increased by well more than
100 percent, moving from $6 per short ton in March 2003 to about $15 per short ton in March 2006. As
more long-term contracts that dominate the average delivered coal prices in Figure 2-8 begin to expire, the
effect of this dramatic increase in spot coal prices will begin to emerge as fuel price increases for coal-fired
generators.
Figure 2-8
Historical Delivered Coal Prices ($ Nominal)
2.00
1.75
$ / MMBtu
1.50
1.25
1.00
0.75
0.50
0.25
Year
Source: EIA Monthly Energy Review.
Figure 2-9
Coal Spot Prices (May 2003 to May 2006)
Source: EIA - http://www.eia.doe.gov/cneaf/coal/page/coalnews/coalmar.html#spot.
16
2005
2003
2001
1999
1997
1995
1993
1991
1989
1987
1985
1983
1981
1979
1977
1975
1973
0.00
Why Are Electricity Prices Increasing? An Industry-Wide Perspective
Various considerations have influenced the rise in coal prices over time. First, high natural gas prices have
shifted some demand from gas to coal, while rising oil prices have driven up the costs of mining and
shipping coal. Second, the effects of Hurricane Katrina also caused some disruptions that put upward
pressure on coal prices, although the primary issues affecting prices recently have been transportation costs
and disruptions. Two major train derailments created severe disruptions for delivery from the Powder River
Basin. Initial repairs were completed in late 2005, but prices increased during the meantime as alternative
transportation routes were developed and Union Pacific suspended new southern Powder River Basin
business.12
Beyond the commodity cost of coal itself, two other important factors influence the cost of coal generation.
As mentioned above, transportation costs are critical to the coal industry. While spot prices reflect
transportation-related shocks, such as the Powder River Basin derailments, they do not directly reflect the
price of transportation, which is growing with overall energy prices. Another driver is recently high sulfur
dioxide (SO2) emissions allowance costs in the SO2 permit trading market.
Figure 2-10 plots monthly SO2 and nitrogen oxides (NOx) prices for the past several years. This figure
shows the exponential growth in SO2 spot prices from below $200 per ton in 2003 to record high levels of
nearly $1600 per ton in late 2005. SO2 emissions allowance prices have since retreated from their highs at
the end of 2005, with May 2006 data showing average prices at $606 per ton of SO2, about triple the price
levels experienced between 2000 and 2003. This allowance price more closely reflects a level that balances
the costs and benefits for the installation of scrubbers while utilities face tightened emission caps. If
emissions allowance prices fall below this level, the cost of installing scrubbers may begin to exceed the
benefits from avoiding the purchase of allowances. Conversely, as the allowance prices rise above this level,
the industry will have further incentives to build additional scrubbing capacity. Thus, although spikes such
as those observed in 2005 may be transitory, the industry has begun to respond by installing environmental
controls on more units to reduce reliance on emission allowances that will become increasingly scarce. As
allowance allocations shrink in the future and generators install controls on smaller, more expensive units,
allowance prices may gradually rise in the future as current federal and state clean air regulations are
implemented and new programs are enacted.
12
http://www.eia.doe.gov/cneaf/coal/page/special/feature.html.
17
Chapter 2: Increased Fuel Prices Drive Utility Costs
Figure 2-10
SO2 and NOx Emissions Allowance Prices
9,000
8,000
1,400
7,000
1,200
6,000
SO2 $/ton
1,600
1,000
5,000
800
4,000
600
NOX $/ton
SO2 and NOX Emissions Allowance Prices
1,800
3,000
SO2 Emissions
Allowance Price
400
NOX Emissions
Allowance Price
200
1,000
-
Jun-00
Sep-00
Dec-00
Mar-01
Jun-01
Sep-01
Dec-01
Mar-02
Jun-02
Sep-02
Dec-02
Mar-03
Jun-03
Sep-03
Dec-03
Mar-04
Jun-04
Sep-04
Dec-04
Mar-05
Jun-05
Sep-05
Dec-05
Mar-06
-
2,000
Source: Argus Air Daily
Monthly Average Prices.
Month
With the expected lag between spot price movements and ultimate delivered prices to electric generators,
prices should continue to escalate in the near term as new contracts begin to reflect the commodity cost of
coal. Figure 2-11 provides a variety of nominal dollar coal price forecasts for deliveries to electric
generators. The lines reflect EIA Annual Energy Outlook 2006 Reference, High Economic Growth, and Low
Economic Growth scenarios. EIA expects the growth rate of coal prices to outpace inflation until about
2007, and then expects a slower level of growth in nominal coal prices.
More variability exists among longer-term forecasts. For example, Energy Ventures Analysis13 predicts that
coal prices will be significantly higher than today in nominal terms, meaning that recent market events are
expected to remain locked into prices. Conversely, Global Insight14 projects much lower coal prices, below
the bottom end of EIA's range. While all signs point to continued near-term increases in the delivered price
of coal, the range in longer-term forecasts reflects uncertainty related to this critical input cost.
13
14
Published in August 2005.
Published in Summer 2005.
18
Why Are Electricity Prices Increasing? An Industry-Wide Perspective
Figure 2-11
Forecasts of Delivered Coal Prices ($ Nominal)
2.00
Energy Ventures
Analysis Forecast
1.90
EIA Reference, High
and Low Economic
Growth Cases
$ / MMBtu
1.80
1.70
Global Insight,
Inc. Forecast
1.60
1.50
Historical
Data
1.40
1.30
2015
2014
2013
2012
2011
2010
2009
2008
2007
2006
2005
2004
2003
1.20
Year
Sources and Notes:
EIA Annual Energy Outlook 2006.
Real prices converted to nominal using forecast GDP deflator.
Nuclear Fuel
Although the fuel component for nuclear energy is a relatively small portion of operating costs compared to
fossil fuel-fired generation,15 the price of nuclear fuel has risen recently as well. Figure 2-12 displays the
historical weighted average price of milled uranium (U3O8) purchased by owners and operators of U.S.
civilian-owned and operated nuclear reactors. Before analyzing the trends presented in this figure, it is
important to note that additional costs are incurred before the milled uranium (or yellowcake) is useful for
power generation. Conversion to uranium hexafluoride, UF6, and subsequent enrichment to increase the
concentration of the fissionable isotope involve energy-intensive processes that become more expensive as
energy prices increase, and then the enriched UF6 is converted into nuclear fuel.16 Each of the production
steps represents additional costs not captured in the wholesale purchased price of milled uranium.
The market for milled uranium has experienced price increases that track the direction of increases in fossil
fuel costs. Between 2001 and 2005, wholesale prices for milled uranium increased from $10.15 per pound of
U3O8 to $14.36 per pound, an increase of about 40 percent. Uranium is purchased largely from foreign
suppliers: in 2005, 60 percent of total purchased uranium came from abroad. Purchase prices from foreign
suppliers rose nearly 50 percent between 2001 and 2005, an increase that exceeded that of the weighted
average price of uranium over this period.
15
A somewhat dated estimate finds that nuclear fuel costs amount to only less than one half of one cent per kilowatt-hour.
See: http://www.eia.doe.gov/cneaf/nuclear/page/analysis/nuclearpower.html.
16
http://www.eia.doe.gov/cneaf/nuclear/page/intro.html.
19
Chapter 2: Increased Fuel Prices Drive Utility Costs
Figure 2-12
Weighted Average Purchased Uranium Price ($ Nominal)
16
$ / pound U3O8 Equivalent
14
12
10
8
6
4
2
2005
2004
2003
2002
2001
2000
1999
1998
1997
1996
1995
1994
0
Source: EIA.
Purchased Power Costs
Many utilities—especially those in states that have undertaken restructuring efforts—rely heavily on power
purchased on the wholesale markets to fulfill their load-serving obligations. As a result of the fuel price
increases cited previously, these purchased power costs have risen dramatically in the past two years.
Wholesale power prices have responded to the marginal fuel prices—primarily natural gas in the peak
periods and coal in the off-peak hours—in ways that have amplified the impact of fuel price increases.
Before analyzing price trends in purchased power costs, it is informative to review the evolution in wholesale
power markets.
Prior to 1990, almost all of the power transacted in wholesale markets was sold at cost-based rates. Around
1990, the Federal Energy Regulatory Commission (FERC) began to permit wholesale power providers,
including vertically integrated utilities, to sell at market-based rates as long as the seller showed that it did
not have market power or, if it did, that it had sufficiently mitigated such market power.17 FERC also
approved regional “standardized” tariffs, such as the Western Systems Power Pool Agreement, that permitted
all members of the Agreement to sell power among themselves without having to file transaction-specific or
bilateral agreements. The Energy Policy Act of 1992 encouraged the trend toward market-based pricing by
creating a class of generators known as Exempt Wholesale Generators (EWGs). Such generators, also
commonly known as Independent Power Producers (IPPs), were permitted to sell in wholesale markets at
unregulated rates. As the 1990s progressed, sales at market-based rates became common in wholesale power
markets.
17
See, for example, Boston Edison Company Re: Edgar Electric Energy Company, 55 FERC ¶ 61,382 (1991), and Heartland
Energy Services Inc., 68 FERC ¶ 61,223 (1994).
20
Why Are Electricity Prices Increasing? An Industry-Wide Perspective
FERC later approved the formation of centralized power markets and power exchanges in which hourly spot
energy prices (both day-ahead and real-time) are set by an Independent System Operator (ISO) or Regional
Transmission Organization (RTO). In such markets, generators bid the price at which they are willing to sell
power. The market-clearing price is the price of the last unit needed in a given hour to serve load. Hence, in
formal wholesale power markets, hourly prices are set by the interplay of demand and supply rather than by
any particular seller’s average cost of service. Whereas cost-based rates tend to be stable because a seller’s
average costs do not change significantly, particularly in the near-term, market prices can fluctuate
significantly from hour to hour based on sudden changes in demand, generation unit availability, and
transmission constraints, among other factors. All of these factors can lead to a significant increase or
decrease in the marginal bid accepted by the RTO to serve demand.
Today, there are five centralized energy markets in the United States—the markets operated by the
California ISO, the Midwest ISO, PJM, New York ISO, and ISO New England.18 The Southwest Power
Pool is implementing a real-time energy market that will have some of the attributes of the centralized
markets cited above. Texas, via its statewide reliability council (ERCOT), has initiated the development of a
centralized nodal market by the beginning of 2009.19
Market-based power pricing also is common in regions without centralized energy markets. In such regions,
power generally is traded through bilateral contracts that range in length from one day to several years. As
in the centralized markets, however, wholesale prices are affected by changes in market fundamentals (e.g.,
demand, unit availability, fuel costs, and transmission bottlenecks), and thus can fluctuate significantly on a
day-to-day basis and over time. Prices in such markets will be stable only insofar as the underlying market
conditions are stable.
Wholesale Prices Are Increasing and Becoming More Volatile
Figure 2-13 demonstrates the upward trend in average daily energy prices in centralized energy markets.
Shown are monthly and daily energy prices in PJM, New York ISO, ISO New England, and the Midwest
ISO over as much of 2001 to 2005 as the markets operated.20 A more limited set of observations for the
Midwest ISO market is available because it opened in April 2005.
As one can see, average prices have risen on a gradual upward trajectory since 2002. These averages of
daily prices have varied from about $21 per MWh to more than $116 per MWh in the eastern power markets.
While certain seasonal patterns are predictable, the price levels themselves have varied significantly from
year to year and over shorter periods as well. Most important, during 2005 spot prices on nearly all markets
rose by almost 100 percent, mirroring the increases in fuel costs just discussed.
18
The California ISO only has a real-time energy market, whereas the other four markets have a real-time and a day-ahead energy market.
http://www.ercot.com/news/press_releases/2006/ERCOT_at_a_Glance_News_Update__February_9%2C_2006.html#Fee%20Case%20Hearing.
20
We purposely have excluded prices from California and the western United States because the 2000-2001 western power
crisis was a highly unusual and unprecedented episode that FERC has determined was caused in part by market
manipulation.
19
21
Chapter 2: Increased Fuel Prices Drive Utility Costs
Figure 2-13
Average Day-Ahead Energy Prices (2001 to 2005)
140
PJM ISO
120
New York ISO
$ / MWh
100
New England ISO
80
Midwest ISO
60
40
20
9/2005
5/2005
1/2005
9/2004
5/2004
1/2004
9/2003
5/2003
1/2003
9/2002
5/2002
1/2002
9/2001
5/2001
1/2001
0
Month
Sources and Notes:
Global Energy Decisions. Prices represent mean LMP prices within each market.
Midwest ISO begins in April 2005. New England ISO begins in March 2003.
Bilateral energy markets also have experienced upward trends that track increases in fuel prices. Figure 2-14
shows monthly energy prices at popular trading hubs in the Midwest, Southeast, and West. The prices are
calculated as a volume weighted average of daily prices. The upward trajectory in prices experienced at all
of these trading hubs is comparable to the pattern observed in the centralized power markets. In particular,
2005 saw the same approximate doubling of prices over the course of a single year.
Figure 2-14
Average Daily Bilateral Energy Prices at Major Hubs by Month
j
y
150
Midwest - Cinergy
120
$ / MWh
Southeast - Entergy
90
West - Palo Verde
60
30
9/2005
5/2005
1/2005
9/2004
5/2004
1/2004
9/2003
5/2003
1/2003
9/2002
5/2002
1/2002
0
Daily Average by Month
Sources and Notes:
Global Energy Decisions.
The Midwest region was organized into the Midwest ISO in April 2005, and for this reason the datasource for Cinergy
changed between March and April 2005.
22
Why Are Electricity Prices Increasing? An Industry-Wide Perspective
Price volatility is expected in commodity markets and can provide market participants useful short-term
signals. Electricity markets are particularly volatile because, unlike most other commodities, electricity
cannot be stored and its short-run demand is highly price inelastic. This makes electricity prices particularly
sensitive to sudden changes in market conditions, such as the loss of a large generating plant or large
transmission line, or large shocks in input costs. While the preceding two figures demonstrate that price
levels themselves have increased, they do not document volatility in electricity prices, per se.
Figure 2-15 presents a crude measure of volatility, by plotting the standard deviation of daily prices for each
month for the hubs presented in Figure 2-14. Clearly, bilateral prices have faced periods of significant
volatility over the time period, and currently reflect significant variation within each month. In response to
the fuel price increases documented in this chapter, wholesale volatility is now higher than at any prior time
except for a brief period of energy price spikes in 2003. This trend translates into higher purchased power
costs, which dominate utilities’ core operating expenses.
Figure 2-15
Standard Deviation of Daily Bilateral Energy Prices
At Major Hubs by Month
30
Midwest - Cinergy
25
Southeast - Entergy
20
West - Palo Verde
15
10
5
9/2005
5/2005
1/2005
9/2004
5/2004
1/2004
9/2003
5/2003
1/2003
9/2002
5/2002
0
1/2002
Standard Deviation ($ / MWh)
35
Average Daily by Month
Sources and Notes:
Global Energy Decisions.
The Midwest region was organized into the Midwest ISO in April 2005, and for this reason the datasource for Cinergy
changed between March and April 2005.
23
CHAPTER 3
Drivers of Electricity Demand
Increasing Demand for Power
Increasing demands for electricity are a fact of life in the American economy. When demand increases, both
utility obligations and market prices signal the need for new investments in generation and power delivery.
This phenomenon occurs even as the cost of providing electric service increases.
In this chapter, we begin by noting the longstanding relationship between economic growth, technical
progress, and the increased electrification of the economy. We then examine the potential impact of higher
electricity prices on demand growth, and the prospects of demand-side conservation programs to slow
demand growth. These considerations combine to paint a picture of the need for new power industry
investments required for adequate and reliable service.
Academic economists generally agree that new knowledge and innovation likely account for 80 percent to 90
percent of total factor productivity growth. In turn, productivity growth is estimated to account for more
than half of GDP growth. Consistent with these results, Boskin and Lau (1992) found that during the period
from 1948 to 1985, “technical progress accounted for half or more of an industrialized nation’s economic
growth.”21
Some economists have found an important causal link between electrification and technological progress.
For example, Sam Schurr found that, during much of the 20th century, technological advance has been
energy dependent, which means that during this era of rapid productivity growth there also was a substantial
increase in the ratio of energy used to the quantities of labor and capital. Moreover, Schurr concluded that in
the middle of the 20th century, there was a major transition to the use of electricity. In his view, this latter
development, in particular, helped to increase the overall flexibility of production, thereby leading to the
growth of economic productivity.22
Technical progress is certainly related to electricity-based innovation, which can create opportunities for
productivity growth in two ways: (1) developing improved electric end-use technologies and (2) improving
the ultimate efficiency of the electricity infrastructure itself. The first growth factor can be expected to
21
22
Electric Power Research Institute, Electricity Technology Roadmap 1999, Summary and Synthesis, p. 53.
Sam H. Schurr, Electricity Use, Productive Efficiency and Economic Growth: A Workshop, Electric Power Research
Institute, 1986, p. 3.
25
Chapter 3: Drivers of Electricity Demand
increase demand for electricity. Technological progress that contributes to end-use efficiency, however,
reflects part of the second driver of demand and has an opposite effect.
Some new technologies and processes, commonly known as electric solutions or electrotechnologies,
substitute electricity for energy traditionally supplied by fuel combustion, raising overall electricity use. In
many cases, these electric applications are themselves efficient enough that they actually use less energy
overall, even accounting for the fuel used to generate the electricity (i.e., they actually reduce primary energy
use). The following text box provides some examples of such technologies.
Electric Solutions
Some technologies and processes, commonly known as electric solutions, increase the use of
electricity while reducing overall primary energy consumption. These technologies foster enduse efficiency and reduced environmental impact. Some technology advances will improve the
efficiency of existing applications, such as high-efficiency lighting and motors. Other electric
solutions will replace existing fossil-fueled equipment but operate at a higher efficiency, as in
the eventual substitution of plug-in hybrid vehicles (which are charged directly from the
electric grid) for gasoline-powered cars, and the use of high-efficiency heat pumps, such as
geothermal heat pumps, for home and commercial building applications.
Finally, some electric solutions introduce completely new processes to improve both energy
efficiency and productivity simultaneously. An example is microwave synthesis of ethylene,
which replaces a chemically driven cracking process with a microwave process that consumes
far less energy by breaking and forming only the chemical bonds required to complete the
reaction. Moreover, it creates a much smaller and less hazardous waste stream. Another
example is isothermal melting (ITM), a system that uses immersed electric heaters to melt
metal by heat conduction. ITM melts metal at a much lower temperature than traditional gasfired furnaces that use heat radiation. As a result, ITM uses far less energy than traditional
furnace technology. With a 60-percent market penetration in 2020, ITM would save
approximately 18.6 trillion Btu and reduce emissions by more than 180,000 metric tons of
carbon equivalent.2
1
Electric Power Research Institute, Electricity Technology Roadmap: 1999 Synthesis and Summary, p. 86.
2
http://www.eere.energy.gov/industry/aluminum/pdfs/itm_1_7.pdf.
Greater electrification in a growing economy can be observed in EIA's projected electricity use per square
foot of commercial sector capacity. (See Figure 3-1.) While EIA reports that the growth rates for overall
energy use and commercial space expansion are quite similar, this chart reveals that electricity use is
26
Why Are Electricity Prices Increasing? An Industry-Wide Perspective
expected to increase steadily in commercial establishments. EIA indicates that electric power intensity will
increase as these establishments add linkages to the Internet and other telecommunications options.23
26
22
18
14
Source:EIAEIA
AEO
2006.Outlook 2006.
Source:
Annual
Energy
2029
2027
2025
2023
2021
2019
2017
2015
2013
2011
2009
2007
2005
10
2003
Total annual KWh per Square foot
Total annual kWh per square foot
Figure 3-1
Commercial Electricity Use Per Square Foot of Capacity
Year
A wide variety of projections expect steady growth in the demand for electricity for the foreseeable future.
Figure 3-2 displays historical demand for electricity and a variety of projections from EIA. The chart also
displays NERC's most recent projections for net electric energy for load, built up from individual NERC
subregion reports. The figure shows that EIA projects 14-percent growth in U.S. demand between 2006 and
2014 under its reference case scenario, and 11-percent to 17-percent growth for the same time period under
its low and high macroeconomic growth scenarios, respectively.24 NERC projects 16-percent growth in net
energy for load and 17-percent peak demand growth for the 2006 to 2014 time period, equivalent to EIA’s
High Growth Case.
23
24
http://www.eia.doe.gov/oiaf/aeo/pdf/trend_2.pdf.
EIA also ran a sensitivity with high and low global petroleum prices, and these results showed a range of growth in
demand for the same period from 13 percent to 16 percent for the high oil price and low oil price scenarios.
27
Chapter 3: Drivers of Electricity Demand
Figure 3-2
U.S. Electricity Demand (1985 to 2014)
5000
EIA - High Growth Case
4500
EIA - Reference Case
BillionKWh
kWh
Billion
EIA - Low Growth Case
4000
EIA - Reference Case
NERC - Net Electric
Energy for Load
3500
EIA - High Growth Case
EIA - Low Growth Case
3000
EIA - Annual Energy Review
2500
NERC - Net Electric Energy
for Load
Sources and Notes:
EIA Annual Energy Outlook 2006.
NERC ES&D report.
2013
2011
2009
2007
2005
2003
2001
1999
1997
1995
1993
1991
1989
1985
2000
1987
EIA - Annual Energy
R i
Year
`
As many analysts have observed, since 1970 the real U.S. economy has grown by nearly 200 percent while
energy consumption has increased by only about 50 percent. This “decoupling” of energy and GDP has
occurred in part because the U.S. economy became more energy efficient overall and also because the
composition of our GDP has shifted away from energy-intensive products toward services.25
Part of the U.S. economy's increase in energy efficiency coincides with a shift toward the greater use of
electric power, which tends to perform some tasks more efficiently than other fuels. In 1950, 14 percent of
energy consumed in America was used to produce electricity. By 1970, that fraction increased to 24 percent,
and today electricity accounts for 39 percent of total primary energy usage.26 In particular, the
miniaturization and digitalization of many technologies, as well as increased generator efficiencies, are
reducing energy use as electric power demand continues to rise.
These relationships are illustrated in Figure 3-3. In the late 1980s and early 1990s, output and electricity use
tended to grow at the same rate. Starting in the late 1990s, economic output grew faster than electricity use
and much more rapidly than overall energy consumption.
25
It should be noted that these figures reflect “direct energy,” or energy consumed in the United States. The energy
consumed to make products that are imported (“indirect energy”) is not reflected in these figures.
26
Energy Information Administration, Annual Energy Review 2004, Table 2.1a.
28
Why Are Electricity Prices Increasing? An Industry-Wide Perspective
Figure 3-3
Indices of Electricity Use, Energy Use, Real GDP
y
gy
Real
GDP
(+79%)
180
170
Electricity
Use
(+60%)
Index (1985=100)
160
150
140
Energy
Use
(+30%)
130
120
110
100
2003
2001
1999
1997
1995
1993
1991
1989
1987
1985
90
Year
Sources and
andNotes:
Notes:EIA Annual Review 2004
Sources
EIA Annual Energy Review, 2004
Looking toward the future, these two overall trends are expected to continue. Figure 3-4 shows a ratio of
total projected U.S. electricity use to total projected GDP in EIA’s latest long-term forecast. This forecast
shows that electricity consumption per dollar of GDP is expected to drop by more than 25 percent over the
next 20 years. While such efficiency gains are expected in every sector, we examine the trends more closely
in the household sector in Appendix A.
Figure 3-4
Consumption of Direct Energy and Electricity vs. GDP
In EIA Long-Term Forecast
g
1.50
Index, 2004 = 1
Index (2004 = 1)
1.25
1.00
0.75
Energy Consumption per $ of GDP
0.50
Electricity Consumption per $ of GDP
0.25
Source: EIA AEO 2006.
2030
2028
2026
2024
2022
2020
2018
2016
2014
2012
2010
2008
2006
2004
0.00
Year
Source: EIA Annual Energy Outlook 2006.
29
Chapter 3: Drivers of Electricity Demand
The Effect of Price Increases on Power Demand
Like all other normal goods, economic research and industry experience have confirmed that an increase in
the real price of electricity will lead to a reduction in the growth of power demand. Because the electric
industry is likely to go through a period of real price increases, it is important to examine the extent to which
expected rate increases might reduce future expected demand growth, which influences utility forecasts of
required investment.
Earlier in this chapter, Figure 3-2 presented EIA and NERC projections of increased power demand through
2014. While we do not have the data underlying NERC’s projection, we can investigate the extent of
potential price response effects using EIA’s publicly available input data. To gain a better sense of the
potential magnitude of the price response of demand, we have conducted a simple sensitivity calculation of
possible price effects on EIA's projections. Details of the calculations are discussed in Appendix B.
Our sensitivity analysis, illustrated in Figure 3-5, examines the impact of a hypothetical price increase that
differs from the EIA projection. In this figure, the dark blue line indicates EIA’s original reference case
projections of real retail prices, which EIA expects to decline after the peak observed in 2005. The solid
black line shows the sum of electricity demand in the residential, commercial, and industrial sectors in this
same reference case. Note that in response to the nearly 10-percent one-year rise in prices between 2004 and
2005, demand flattens out considerably between 2005 and 2006.
Our sensitivity analysis simply calculates the potential effect of a sustained real power price increase, using
the same short-run EIA price response (elasticity) assumptions incorporated into its forecasting model. In
particular, real prices are assumed to increase 10 percent between 2005 and 2006, and then no change in real
price is assumed through 2014. Demand is then adjusted based upon the short-run elasticity factors from
EIA and the difference in price from the underlying forecast in a given year.
The resulting demand growth projection is the dashed black line in Figure 3-5, where the blue shaded area in
the figure reflects the loss in demand in response to the hypothetical higher prices. In this simple
experiment, an increase in the projected real price of electricity reduces overall demand growth in the 2006
to 2014 period from 14.5 percent to 10.6 percent. Put another way, approximately 175 billion kilowatt-hours
(kWh) of expected demand in 2014 would not be realized due to the price response of demand in this
illustrative analysis. Using EIA’s projected capacity factors for coal generators by 2014, this is equivalent to
obviating the need for about 25 gigawatts (GW) of coal-fired capacity in 2014.
30
Why Are Electricity Prices Increasing? An Industry-Wide Perspective
Figure 3-5
Hypothetical Response of Demand to Change
In Real Electricity Prices
9
8
4800
4600
EIA Projected Retail Prices
7
4400
6
4200
EIA Projected Demand
5
4000
4
Estimated
Price
Response
of Demand
3800
Hypothetical Demand Given
Permanent 10% Real Price
Increase in 2006
3
2
3600
3400
2014
2013
2012
2011
2010
2009
2008
2007
3000
2006
0
2005
3200
2004
1
Annual Consumption (billion kWh)
5000
Hypothetical Price Series Assuming Permanent 10%
I
i 2006
2003
Electric Rates (2004 cents / kWh)
10
Source: EIA Annual Energy Outlook. Reflects demand and prices for residential, commercial, and industrial electricity.
This calculation is only illustrative, but it highlights the linkage between prices and forecast demand that can
have a substantial impact on the amount and timing of new generating capacity needed. In addition to the
demand response to increased electricity prices, demand is likely to be moderated through an expansion of
demand-side management and demand-response programs adopted by utilities.
The Impact of Demand-Reduction Programs
The need for additional generation and transmission capacity will be mitigated somewhat by demand and
energy reductions achieved through a variety of conservation, energy efficiency, and demand-response (DR)
or load-management programs.27 Such programs realize demand and energy savings in addition to those
achieved through the price elasticity effects discussed earlier. It is important to recognize that customers, not
utilities, control their own usage of electricity, except for those demand management programs that involve
interruptible service. Conservation and energy efficiency programs reduce customers’ overall electricity
consumption, whereas DR programs reduce customers’ consumption during peak hours but do not
necessarily reduce their overall electricity consumption. Examples of the former used in the past include
financial incentives or rebates to encourage customers to buy more efficient appliances (refrigerators, water
heaters) while examples of the latter include programs that enable a utility to temporarily shut off a
customer’s air conditioner or water heater during high demand periods. Both types of programs came to be
known as demand-side management (DSM) programs.
27
We use the latter two terms interchangeably in this report.
31
Chapter 3: Drivers of Electricity Demand
Many utilities, at the behest of their state regulators, started to implement conservation and load management
programs during the early 1980s largely as a response to rising fuel costs, increasing construction costs, and
growing public concerns about the environmental impacts of fossil-fired and nuclear generation. These
programs sought to educate and motivate customers into adopting more efficient appliances, or allowing
utilities to cycle or shut off end-use equipment for short periods during peak conditions. In addition, many
regulators and stakeholders came to believe that demand-side resources—both energy efficiency and load
management—should be considered and evaluated in utility resource plans in an integrated fashion with
traditional supply-side resource options. Indeed, the need to consider demand-side and supply-side resources
in an integrated manner was a central tenet of what came to be known as integrated resource planning (IRP).
By 1991, many state commissions had implemented some form of IRP, though the extent to which utilities
were required to pursue DSM programs varied widely.28
Most utility DSM programs were scaled back in the mid-1990s as retail rates (in real terms) declined and
states implemented or considered retail competition. According to EIA, 1993 was the high water mark for
utility spending on DSM programs, with total nationwide expenditures of more than $2.7 billion (including
both direct and indirect program costs). By 1996, total utility DSM spending had fallen to $1.9 billion and
by 1999 it had fallen to less than $1.5 billion. In many restructured states, however, DSM spending authority
shifted from utilities to state or non-profit entities (through the collection of system benefit charges or public
benefit funds), expenditures that are not reflected in these figures. Figure 3-6 shows the trend in utility DSM
spending over the 1989 to 2004 period, normalized by expressing DSM costs as a percentage of retail sales.
Figure 3-6
Normalized Utility Demand-Side Management
Program Costs, 1990 Through 2004
retail sales
DSM costs as a percent of
DSM Costs as a percent of
retail sales
1.6%
1.4%
1.2%
1.0%
0.8%
0.6%
0.4%
0.2%
2004
2002
2000
1998
1996
1994
1992
1990
0.0%
Year
Sources and Notes:
Form EIA-861, "Annual Electric Power Industry Report".
This graph tracks only the DSM spending made by utilities, and does not include spending by state or non-profit entities in
restructured states
28
Cynthia Mitchell, “Integrated Resource Planning Survey: Where the States Stand,” Electricity Journal, May 1992,
pp. 10-15.
32
Why Are Electricity Prices Increasing? An Industry-Wide Perspective
The recent increases in fuel and power prices have spurred renewed interest in utility-sponsored DSM
programs. In September 2005, the California Public Utilities Commission approved $2 billion in funding for
energy-efficiency programs from 2006 through 2008, an effort state regulators called the most ambitious
energy-efficiency and conservation campaign in the history of the United States. These efficiency programs
are expected to cut energy costs by more than $5 billion and eliminate the need to build three large power
plants over the next three years.29 In late 2005, the Energy Center of Wisconsin prepared a report that
concluded, over the next five years, an average of $75 million to up to $121 million per year could be spent
cost-effectively on statewide programs aimed at improving energy efficiency in Wisconsin homes and
businesses. In Arizona, regulators mandated $48 million in new efficiency programs in 2005 and recently
ordered an additional $21 million in spending. Clearly, interest in, and expectations from, DSM programs
are increasing as states grapple with increased costs and rising prices.
Historical Energy and Demand Savings from DSM Programs
EIA also collects data on the energy and demand savings achieved by utility DSM programs. These savings
have been relatively consistent over the 1994 to 2004 period, which suggest that DSM programs initiated in
the early 1990s have produced relatively consistent savings over the last 10 years. For example, EIA found
that in 1994 DSM programs saved a total of 57,421 GWh of energy, which is equivalent to the annual output
of seven large nuclear units or the annual output of about 20 500-MW-generating units operating at a 66percent capacity factor. In 2004, these programs saved 54,710 GWh of energy.
DSM programs also reduce peak load. According to the EIA data, DSM programs have reduced peak load
by at least 23 GW over the 1994 to 2004 period. Peak load reductions have been relatively consistent over
this period, ranging from a low of 22.9 GW in 2000 to a high of 29.8 GW in 1996. (Peak demand savings
will be more sensitive to weather than energy savings and therefore somewhat more likely to vary from year
to year.) In 2004, peak load reductions were 23.5 GW, a significant savings—a typical new combustion
turbine (CT) is about 100 MW, so existing DSM programs have displaced the need for more than 200 CTs
nationwide. Approximately 60 percent (14.3 GW) of the demand reduction savings were achieved by
energy-efficiency programs, with the remainder attained through load-management programs.
Potential Energy Savings from DSM Programs
Several studies recently have been conducted on the technical, economic, and/or achievable potential for
energy efficiency in the United States These studies evaluated the potential for saving electricity, natural
gas, or both, within a specific state or region, through various conservation measures and programs. Nadel et
al. reviewed and compared these studies to reach preliminary conclusions about the level of achievable
energy efficiency in the United States.30
29
30
“California PUC OKs $2B for Energy Efficiency,” Megawatt Daily, September 23, 2005.
Steven Nadel, Anna Shipley, and R. Neal Elliott, The Technical, Economic, and Achievable Potential for EnergyEfficiency in the U.S. – A Meta-Analysis of Recent Studies, Proceedings of the 2004 ACEEE Summer Study on Energy
Efficiency in Buildings, 2004.
33
Chapter 3: Drivers of Electricity Demand
Eight of the studies reviewed by Nadel examined potential electricity savings. A subset of these studies
estimated achievable potential savings, which take into account the rate at which homes and businesses will
actually adopt energy-saving technologies and practices. As such, achievable potential is a more
conservative measure of potential energy-efficiency savings than either technical potential or economic
potential.31 Achievable potential savings ranged from 10 percent to 33 percent, with two studies estimating
savings of 10 percent to 11 percent, two estimating savings of 31 percent to 33 percent, and one estimating
24 percent, with the last estimate being the median. This range shows that results are very sensitive to the
study’s underlying assumptions. The median estimate of 24 percent translates into achievable potential
savings of 1.2 percent per year. If realized, these savings would reduce annual electricity growth by
approximately 50 percent, an ambitious but potentially plausible goal if conservation programs were pursued
aggressively across the United States.
It is unknown at this time how aggressively each state will pursue energy-efficiency programs over the next
10 years and how states will address the difficult policy and ratemaking issues that such programs entail.
Spending on conservation programs is likely to increase, but the magnitude and pattern of the increase are
very uncertain and will be set on a state- or utility-specific basis. This, in turn, makes estimated savings very
uncertain, because the studies cited show that achievable savings will be sensitive to the aggressiveness of
the program and policy tools employed by states and utilities.
Savings from Appliance and Equipment Standards
Significant energy savings also are provided by federal appliance and equipment efficiency standards.
Federal standards were first adopted in 1987, through the National Appliance Energy Conservation Act of
1987, and then extended through the Energy Policy Act of 1992 and again through EPAct 2005. The
specific products (from the 1987 and 1992 laws) covered by these different standards are summarized in
Table 3-1. These standards prohibit the production and import or sale of appliances or other energyconsuming products less efficient than the minimum requirements. In addition, the U.S. Department of
Energy (DOE) also establishes building codes that set minimum efficiency levels for appliances and
equipment installed in new homes.
31
However, several studies estimated achievable potential but not economic potential, which means that median results for
economic and achievable potential cannot be directly compared.
34
Why Are Electricity Prices Increasing? An Industry-Wide Perspective
Table 3-1: Products Subject to Existing Appliance Efficiency Standards
Products Included in the National Appliance Energy Conservation Act (NAECA)
Refrigerator-freezers
Freezers
Room air conditioners
Central air conditioners and heat pumps
Residential furnaces and boilers
Residential water heaters
Direct-fired space heaters
Clothes washers
Clothes dryers
Dishwashers
Ranges and ovens
Pool heaters
Fluorescent lamp ballasts
Televisions
Products Added in the Energy Policy Act of 1992
Fluorescent lamps
Incandescent reflector lamps
Electric motors
Packaged air conditioners and heat pumps
Commercial furnaces and boilers
Commercial water heaters
Showerheads
Faucets and aerators
Toilets
Distribution transformers
Small electric motors
High-intensity discharge lamps
According to the American Council for an Energy Efficient Economy (ACEEE), the overall savings from
established appliance and equipment efficiency standards have been quite substantial. As of 2000, appliance
standards had already cut U.S. electricity use by 2.5 percent and U.S. carbon emissions from fossil fuel use
by 1.7 percent.32
The recently enacted standards are projected to result in total electricity savings that would reach 253 billion
kWh and 341 billion kWh per year, or 6.1 percent and 7.0 percent of the projected total U.S. electricity use,
in 2010 and 2020, respectively. These standards also are expected to yield peak load reductions of 66 GW
by 2010, which is 7.5 percent of projected total (non-coincident) U.S. peak demand.33,34 It is important to
note that the EIA forecasts discussed earlier in the chapter already account for the savings provided by
appliance efficiency standards and building codes.
32
Toru Kubo, Harvey Sachs, and Steven Nadel, Opportunities for New Appliance and Equipment Efficiency Standards:
Energy and Economic Savings Beyond Current Standards Programs, American Council for an Energy Efficient Economy,
September 2001, p. 5.
33
Id., p. 5.
34
Projected reductions in energy consumption are based on EIA’s 2006 forecast of electricity consumption. The projected
reduction in 2010 demand is based on NERC’s 2005 forecast of peak demand.
35
Chapter 3: Drivers of Electricity Demand
EPAct 2005 mandated national efficiency standards (or rulemaking deadlines) for 16 additional products or
classes of products, including commercial equipment, such as commercial refrigerators and freezers, and
residential equipment, such as ceiling fans and dehumidifiers.35 ACEEE’s preliminary analysis of the impact
of these expanded federal appliance standards finds that they could save an additional 18 billion kWh of
electricity by 2010.36
EPA ENERGY STAR® Program
Another federal initiative that is helping to reduce energy and electricity consumption is the ENERGY STAR
program. The “ENERGY STAR” label identifies products, practices, services, homes, and buildings that
meet government guidelines for energy efficiency. Introduced by the U.S. Environmental Protection Agency
(EPA) in 1992 for energy-efficient computers, the ENERGY STAR program has become a broad platform
for promoting energy efficiency across the residential, commercial, and industrial sectors. The program has
grown to include efficient new homes that became eligible for the ENERGY STAR label in 1995 and more
than 40 product categories for homes and businesses, such as clothes washers, TVs, and refrigerators.37
While the ENERGY STAR initiatives are separate from the utility DSM programs described earlier, EPA in
some cases partners with utilities (as well as home builders, manufacturers, and others who play a key role in
getting energy-efficient equipment into the market).
EPA estimates that the ENERGY STAR programs saved a total of 126 billion kWh of energy and 25 GW of
peak power in 2004—the amount of peak power required for about 25 million homes. These programs also
prevented the greenhouse gas emissions equivalent to those from 20 million vehicles.38
Demand-Response Programs
A variety of DR programs are run by the RTOs and the ISOs in organized markets and by vertically
integrated utilities where traditional industry structures prevail. DR programs can be particularly valuable in
areas that lack sufficient surplus capacity to meet peak demands reliably. Some programs use “price-based”
incentives, such as time-of-use (TOU) or real-time pricing (RTP), to encourage customers to reduce or shift
consumption from peak periods to off-peak periods. Other programs use direct load control devices to
curtail consumption, such as when a utility or system operator remotely shuts down or cycles a customer’s
air conditioner or water heater on short notice to address system or local reliability contingencies.39 In the
PJM market, more than 6,000 commercial and industrial facilities (with peak demand greater than 100 kW),
35
In addition to the appliance standards, EPAct 2005 also includes manufacturer and consumer tax incentives for advanced
energy-saving technologies and practices.
36
http://www.aceee.org/energy/0510confsvg.pdf.
37
U.S. Environmental Protection Agency, Investing in Our Future: Energy Star and Other Voluntary Programs, 2004
Annual Report, October 2005, p. 10.
38
Id., p. 4.
39
Demand-response programs also can recognize the contribution of customer-owned generation (including emergency or
back-up generation) as a means to reduce a customer’s net load, giving appropriate financial value and credit to
participants.
36
Why Are Electricity Prices Increasing? An Industry-Wide Perspective
as well as more than 45,000 small commercial and residential sites, participate in its DR program. ISO New
England currently has more than 600 MW of capacity signed up under its DR program.
Despite these RTO initiatives and utility-run programs, DOE found that limited DR capability exists in the
United States at present. In 2004, the aggregate national DR potential was about 20,500 MW—three percent
of total U.S. peak demand. Actual delivered peak demand reductions were about 9,000 MW, or 1.3 percent
of total peak demand.40 (This is consistent with the findings reported above, in which demand reduction
programs were found to account for about 40 percent of the 23,500 MW reduction in peak load achieved
through DSM programs.) DOE found that the total potential load management capability in the United
States has fallen by 32 percent since 1996 due to low electricity prices, fewer utilities offering load
management services, declining enrollment in existing programs, and the changing role and responsibility of
utilities. DOE also acknowledged some positive developments that suggest a resurgence of interest in load
management. One is the RTO customer load participation programs mentioned. Another is the fact that
some states (Maryland, New Jersey, New York, and Pennsylvania) have adopted real-time pricing (RTP) as
the default pricing mechanism for large customers purchasing generation service from utilities, while others
(California, Florida) have implemented large-scale RTP or critical peak-pricing programs. As a result of
EPAct 2005, time-based tariffs such as RTP will become more prevalent.
Purchases of the advanced controls necessary for some of these DR programs and for greater customer
response to RTP are a valuable industry investment that will reduce power costs in the long run, but will
require an upfront investment by utilities. The text box on the following page describes some of the specific
investments required to better realize the potential of DR.
40
U.S. Department of Energy, Benefits of Demand Response in Electricity Markets and Recommendations for Achieving
Them: A Report to the United States Congress Pursuant to Section 1252 of the Energy Policy Act of 2005, February 2006,
p. xii.
37
Chapter 3: Drivers of Electricity Demand
Investments That Help Realize the Potential of Demand-Response Programs
Four building blocks to improving the efficiency of electricity use are: (1) communications
infrastructure; (2) innovative rates and regulation; (3) smart end-use devices; and (4)
innovative markets. Communications infrastructure is a key enabler of energy efficiency and
demand response because it enables a two-way information exchange between energy service
providers and specific energy-consuming devices. The Internet enables two-way information
exchange with specific end-use devices, provided that the necessary advanced metering
infrastructure is in place.
Smart, network-addressable devices include air conditioners, major appliances, motors,
pumps, and lighting systems. These devices receive electricity rates through the network,
measure and communicate power usage through the network to the energy service provider,
and optimize operation to minimize energy costs. For example, an air conditioner would
measure hourly power consumption and communicate it to the energy service provider
through the Internet.
A number of technologies are available and under development to support demand-response
and energy-efficiency programs. They include a variety of distributed generation technologies
whose costs, with further research and development, are expected to be reduced over time,
thereby enabling their widespread deployment. In addition, several energy storage
technologies are under development that would offer consumers another option for reducing
their electricity demand at peak times when costs are high and reliability may be more likely
to be threatened. Examples of these technologies include:
ƒ Microgrids: Improve power quality, enhance DSM, and ease peak demands resulting
from randomness of load.
ƒ DG and Storage Dispatch, Batteries: Store energy to be used for emergencies or on-peak
needs.
ƒ Super-conducting Magnetic Energy Storage (SMES): Store energy to be used for
emergencies or on-peak needs; real-time control applications.
ƒ Flywheels: Help shave peak demand; enhance power quality and reliability.
ƒ Intelligent Building Systems: Optimize energy consumption.
Advanced meters also will facilitate customer participation in voluntary, price-based demandreduction programs, by allowing all customers to participate in real-time pricing and
comparable programs that provide customers with financial incentives to reduce demand
when production costs or wholesale prices are high. Today, few residential and small
commercial customers have the metering equipment necessary to participate in real-time or
peak-period pricing programs.
38
Why Are Electricity Prices Increasing? An Industry-Wide Perspective
Real-Time Pricing
Under RTP tariffs, electricity customers are charged prices that vary over short time intervals, typically every
hour, and are quoted one day or less in advance to reflect contemporaneous marginal generation costs.41
RTP differs from conventional retail tariffs, which are based on average or embedded costs that typically do
not vary by time of day and are fixed for years at a time. During peak periods, wholesale power costs can be
very high—well above average power costs—but most retail customers receive no price signal indicating
that this is the case. Economists and many policymakers have long believed that RTP or TOU pricing should
be more widely implemented to give customers better price signals and to give them a financial incentive to
reduce electricity consumption at times when wholesale costs are high.42 New computer and fiber
technologies are rapidly being developed that can work within an RTP context to enable utilities and
customers to manage electricity use. However, many customers have resisted RTP because it raises their
peak electricity prices.
Widespread participation in RTP will require increased customer interest in the potential economic benefits
to overcome reluctance to be exposed to wholesale market price fluctuations. High customer participation,
in turn, would entail greater utility program costs for software and technology. For example, RTP will
require replacement of today’s metering technology. Customers who participate in RTP tariffs require
advanced or “smart” meters that measure and store energy usage at intervals of one hour or less and include
communication links that allow the utility to remotely retrieve current usage information whenever needed.
Conventional electro-mechanical meters account for more than 90 percent of the current meter population,43
and only record cumulative energy usage. As noted in Chapter 6, replacement of all conventional meters
with advanced meters would involve a total investment of approximately $12 billion to $18 billion, which
does not include additional equipment needed for RTP, which could substantially increase these costs.44
Thus, while increased use of RTP and other forms of time-based pricing is likely over the next 10 years, it is
unclear how widespread such pricing will become. The push for RTP (and other forms of demand
management) is likely to be strongest in regions with relatively high marginal generation costs, such as New
England and California. Attaining widespread customer participation in RTP will be a challenge, but a
strong push from state regulators could help spur interest even among small customers.
41
Galen Barbose and Charles Goldman, A Survey of Utility Experience with Real Time Pricing, Ernest Orlando Lawrence
Berkeley National Laboratory, LBNL-54238, December 2004, ES-1.
42
See, for example, Kenneth Gordon and Wayne P. Olson, Retail Cost Recovery and Rate Design, Prepared for the Edison
Electric Institute, December 2004.
43
U.S. Department of Energy, Benefits of Demand Response in Electricity Markets and Recommendations for Achieving
Them: A Report to the United States Congress Pursuant to Section 1252 of the Energy Policy Act of 2005, February 2006,
p. 25.
44
This estimate just includes the cost of the meter itself and does not include the cost of the demand-response components,
which vary widely and may be from $100 to $250 per site. Thus, the total cost of providing RTP to the 120 million
residential customers who do not have the necessary technology today could exceed $40 billion.
39
Chapter 3: Drivers of Electricity Demand
Conclusion
The need for additional generation and transmission capacity will be mitigated by demand and energy
reduction achieved through the price elasticity impact of rising prices and through a variety of conservation,
energy efficiency, and demand-response programs. However, there still will be a need in the future for
utilities to make major investments in generation and transmission capacity.
40
CHAPTER 4
Generation Investment
The demand for reliable electricity is expected to grow—even after accounting for greater penetration of
more efficient end-use equipment and potential customer responses to higher prices. In order to meet the
higher demands for more reliable electric service, the industry will need to invest in new generating plants,
transmission facilities, and distribution systems, and will need to make investments in environmental
protection in order to comply with recently enacted regulations.
There currently exists sufficient surplus generating capacity in most regions of the country to meet current
and near-term peak demands reliably, a condition termed “generation adequacy” by utility planners.
However, there are regions, such as the West coast, Florida, and the Northeast, that face more immediate
needs for new capacity to maintain generation adequacy. This chapter examines the likely timing and pattern
of new generation additions in the United States during the next major generation investment cycle. As
explained herein, the most significant changes in the generation investment picture are the re-emergence of
new coal-fired and possibly nuclear baseload generating plants—the first such major additions of solid-fuel
baseload generating plants to the fleet in nearly 20 years—along with significant growth expected in
renewable electric generation. The next wave of generation investments, while not extraordinarily large in
capacity terms in the next decade, marks a turn toward much more capital-intensive types of generation
facilities.
Generation Additions: Past, Present, and Future
The capacity surplus in many regions of the United States is primarily a result of a boom in natural gas-fired
capacity that began in the late 1990s and that is currently winding down as the last few plants are completed
over the next two to three years. This surge in generating capacity is seen quite vividly in Figure 4-1, which
also shows a projection of new capacity and peak demand growth according to NERC.
The reasons for the huge boom in natural gas-fired generation were many, and include:
ƒ Very low capital costs, especially for natural gas-fired combined-cycle (NGCC) plants, owing to rapid
technological improvements;
ƒ Relatively short construction times;
ƒ Very high operating efficiencies and availabilities with low non-fuel O&M expenses;
ƒ Minimal environmental impacts and associated costs;
41
Chapter 4: Generation Investment
ƒ Favorable interest rate environment and low-cost capital structures;
ƒ Optimistically low projections (in retrospect) of natural gas prices.
Figure 4-1
Capacity and Demand Balance
1,000
Capacity / Demand (GW)
NERC ES&D
Generation Capacity
Surge in
Capacity
950
900
850
800
750
NERC ES&D
Peak Demand
700
650
600
Historical
550
Projected
2013
2011
2009
2007
2005
2003
2001
1999
1997
1995
1993
1991
1989
500
Year
Source and Note: NERC Electricity Supply & Demand 2005. Circles reflect forecast values.
For these and other reasons, nearly all generating capacity built since 1995 was natural gas-fired, and a large
portion were NGCC plants whose projected operating economics appeared to be competitive with coal-fired
capacity for baseload and intermediate duty cycles. As shown in Figure 4-2, of the nearly 275 GW of new
capacity installed between 1995 and 2004, more than 260 GW was gas-fired, of which about 135 GW were
combined-cycle plants.
Natural gas prices spiked sharply in 2000, and thereafter increased and became more volatile. The current
price projections discussed in Chapter 2 are two to three times higher in real terms than the prices
experienced in the 1990s. Instead of fuel costs for an NGCC plant in the $25/MWh range (7,000 Btu/kWh
heat rate with $3.50/mmBtu gas prices), these plants now face fuel costs of $75/MWh or more, making them
generally uncompetitive with other baseload technologies such as coal and nuclear. Based on data provided
by Energy Velocity, we calculate that capacity factors for combined-cycle plants have fallen from their highs
of nearly 50 percent in 2001 to 37 percent in 2005.
Thus, while many regions have an installed capacity surplus from a reliability perspective, much of the
surplus arises from plants that are no longer economical to run much of the time, a condition that will persist
if gas prices remain at or near current levels. In some regions, where gas is the marginal fuel for most hours,
the resulting wholesale price increases have helped the economics of operating these plants, but gross
operating margins remain low and the value of these plants has been impaired substantially.
42
Why Are Electricity Prices Increasing? An Industry-Wide Perspective
Figure 4-2
Capacity Additions (1995 to 2004)
80
Capacity Additions (GW)
70
60
50
40
30
20
10
Source: Historical versions of EIA Annual Energy Outlooks
Coal
2004
2003
2002
2001
2000
1999
1998
1997
1996
1995
0
Time Period
Combined Cycle
Nuclear
CTs
Renewables
Low NGCC capacity utilization and high, volatile electricity prices in regions where natural gas influences
wholesale prices have sparked newfound interest in “fuel diversity” at both the federal and state levels. This
is evidenced in Section 1251(a)(12) of EPAct 2005, which requires states to consider means by which they
can minimize their dependence on any one fuel source and to ensure that electricity is produced or procured
from a diverse range of fuels and technologies, including renewables. Minimizing exposure to natural gas
prices also has given a strong impetus to renewable mandates. In the current market, many utilities have
begun to propose non-gas-fired baseload plants to meet future needs. This reflects the widespread perception
that natural gas has become uneconomic and too risky for baseload generation and the corresponding
premise that wholesale electricity market prices now support the construction and operation of new coal and
nuclear plants that economically displace natural gas plant generation.
Although coal and nuclear plants enjoy substantial operating cost advantages over NGCC, they are much
more expensive to build, and face more intensive permitting requirements and longer construction times.
Coupled with the general surplus in most regions, EIA projects that the next generation of baseload plant
construction will begin slowly, with coal and nuclear plant completions accelerating over the next 10 to 20
years, with more renewable capacity in the next 10 years arising from state-level mandates. The EIA
projection of capacity additions is shown in Figure 4-3.
The capacity totals in Figure 4-3 indicate the expected completion dates of new plants, and therefore lag the
bulk of construction expenditures by several years. Thus, the construction expenditures associated with the
plant-in-service projected between 2005 and 2009 already have been partially incurred. The smaller amount
of capacity assumed in service between 2010 and 2014 does not imply a commensurately smaller degree of
capital investment during that period because this is a time when much of the capacity anticipated to be online in the 2015 to 2019 period will be under construction.
43
Chapter 4: Generation Investment
Figure 4-3
Projected Capacity Additions
Capacity Additions (GW)
120
100
80
60
40
20
0
2005 - 2009
2010 - 2014
Source: EIA Annual Energy Outlook 2006.
Coal Steam
2015 - 2019
2020 - 2024
2025 - 2030
Time Period
Combined Cycle
Nuclear
CTs
Renewables
Coal-Fired Generation
Since natural gas prices began rising in the very late 1990s, there have been several coal plants built and
many more proposed. According to DOE, there are currently 140 coal plant proposals that total 85 GW of
capacity and represent a $119-billion investment.45 Of course, these proposals are in various stages, and it is
unlikely that all will be built, or even that the majority will be built within the stated timeframe of 2015
(about 23 GW of the proposed plants do not have a stated in-service date). EIA projects that only about 15
GW of coal-fired capacity will be completed between 2005 and 2014, with another 140 GW between 2015
and 2030.
New coal plants are more efficient and much cleaner than coal plants built 20 years ago, even conventional
pulverized coal technologies. In addition, there has been substantial interest in emerging technologies such
as integrated gasification combined-cycle (IGCC) plants. [The DOE survey tracks 22 proposed IGCC
plants.] IGCC represents a hybrid coal-gas plant, where the coal is gasified under high temperature and
pressure, and the resulting synthetic gas is used to power a combined-cycle plant. While estimates vary,
IGCC construction probably costs about 10 percent to 20 percent more than conventional pulverized coal
plants, and overall efficiency and reliability must be proven beyond the demonstration projects already
completed. Another clean coal technology is circulating fluidized bed (CFB) units, which can also burn
waste coal from abandoned mining sites, with resulting environmental benefits. EPAct 2005 provides tax
credits for early deployments of IGCC (20-percent credit on taxable basis) and other advanced coal
technologies (15-percent credit), subject to national aggregate credit limits for each type of facility.
45
U.S. Department of Energy, Tracking New Coal-Fired Power Plants: Coal’s Resurgence in Electric Power Generation,
March 20, 2006. This includes about 2 GW of coal capacity already in service.
44
Why Are Electricity Prices Increasing? An Industry-Wide Perspective
Investments in coal plants, however, carry significant risks as a result of mandatory controls on greenhouse
gases that might be implemented in the future. The burden of national carbon dioxide (CO2) limits would
fall heavily on coal-fired generation, although older plants are probably at greater risk of closure. However,
any market-based CO2-reduction policy, whether a CO2 tax on fossil fuels or a cap-and-trade allowance
scheme, would significantly raise the operating costs of coal-fired power plants, although the degree of
impact would depend on the particular policy enacted.
Nuclear Power Plants
Interest in building nuclear power plants has revived substantially in the past several years, owing both to
attractive operating economics recently experienced and concerns about global warming policies that might
eventually impair coal investments. In the past 15 years, nuclear power plants have shown tremendous
operational improvements and many have been up-rated to add generating capacity. Average capacity
factors have increased from 66 percent in 1990 to about 90 percent in 2005, owing primarily to increased
availability as refueling outages have been shortened from an average of 104 days to 38 days and to
improved maintenance programs that have reduced forced outages. At the same time, the Nuclear
Regulatory Commission has developed a streamlined licensing process for new nuclear approvals.
Although existing nuclear plants have demonstrated high reliability and very low operating costs, the next
generation of nuclear plants will almost certainly have higher capital costs than conventional fossil fuel units.
However, interest in diversifying the fuel mix and the fact that nuclear power does not emit any CO2 have
led to 10 proposals for new nuclear units, reflecting serious interest in reviving this technology as a baseload
option.46 Some of the project sponsors have already filed for Early Site Permits, and are expected to file for
combined construction and operating licenses within the next two years, which could lead to construction
beginning on some of the plants soon after 2010. EPAct 2005 also encourages new nuclear facilities with a
combination of loan guarantees, production tax credits, and risk protections for initial project developers.
The time horizon for new nuclear investments is such that these investments are not likely to contribute to
upward rate pressures for the period we examine in this paper. However, utilities that are planning these
units will incur some outlays, and future investments at the end of our study period are likely to be
substantial in both size and risk.
46
Fitch Ratings, “Wholesale Power Market Update,” March 13, 2006. Also, Nuclear Energy Institute, “New Nuclear Plant
Status.”
45
Chapter 4: Generation Investment
Renewables
New renewable electricity includes, among others, wind, solar, geothermal, biomass (wood, wood waste,
energy crops, and landfill methane), and small hydro. The primary advantages of renewables are low, stable
operating costs and the environmental benefits of little or no air and water emissions. However, renewable
technologies generally are more costly to build (on an installed $/kW basis), although construction times for
wind and solar are typically shorter than for fossil-based generation capacity. While some biomass and
geothermal operate as baseload capacity, wind and solar have lower capacity factors and their power output
is intermittent because they are based on variable resources. Renewable resources also vary quite
substantially in their geographic distribution.
At this time, wind power is the most competitive renewable generating technology as its levelized cost
compares favorably to the levelized cost of gas-fired generation in some areas. However wind power cannot
reliably meet peak demands because of resource intermittency. Therefore, wind capacity is less valuable to
meeting system reliability and generation adequacy objectives than equivalent amounts of conventional
fossil fuel generation capacity. In addition, variable output that is not readily forecasted makes wind power
more challenging and costly to integrate into the power grid. This additional cost is generally considered
modest at current levels of wind power penetration, but may rise as greater amounts of intermittent resources
are incorporated into regional electricity markets.47 Recognizing the need to promote greater amounts of
intermittent wind resources, FERC and industry stakeholders have developed new market and operational
rules to assist developers in gaining access to transmission and other market services on terms comparable to
those available to conventional energy developers.
Other than traditional hydroelectric power stations, renewable energy is still a small percentage of the overall
electric supply. However, recent growth rates in installed capacity have been impressive—wind capacity has
been growing at about 20 percent per year recently—which has largely been a result of renewable
requirements established at the state level and the periodic renewal of the production tax credit allowed for
renewables, although there also has been increased demand from customers of utilities offering “green”
electricity for a premium rate.
Renewable Energy Standards
A renewable energy standard is a mandate that a retail electricity supplier obtain a specific portion of its total
supply from eligible renewable energy technologies. Most of these standards allow the obligation to be
satisfied by a variety of combinations of renewable sources and are thus referred to as Renewable Portfolio
Standards (RPSs) These policies have been established in 20 states and the District of Columbia as shown in
Table 4-1, and now apply to roughly 50 percent of retail electricity sold. The resources eligible to satisfy the
RPS requirements, the required portion of renewables, and the compliance deadlines vary substantially
among the various state programs. The actual impacts of a given RPS policy on increasing renewable
generation and on the costs of compliance are less related to the absolute percentage requirement but are
more a function of how the actual requirement compares to the eligible renewables already installed and the
47
See Utility Wind Integration Group, “Utility Wind Integration State of the Art,” May 2006.
46
Why Are Electricity Prices Increasing? An Industry-Wide Perspective
potential resource base.48 Thus, a state that already gets a high percentage of generation from renewables
(e.g., Maine) may incur small costs under a nominally ambitious target, while states with smaller percentage
targets but far less potential for economically increasing renewable energy contributions may face very high
costs.
Table 4-1
Fraction of U.S. Retail Load in States with Renewable
Energy Standards
All States with Renewable
Energy Standard
Arizona
California
Colorado
Connecticut
Delaware
District of Columbia
Hawaii
Illinois*
Iowa
Maine
Maryland
Massachusetts
Minnesota**
Montana
Nevada
New Jersey
New Mexico
New York
Pennsylvania
Rhode Island
Texas
Wisconsin
Required Percent
By Year
Credit-trading Policy
in place?
1.10%
20%
10%
10%
10%
11%
20%
8%
2%
30%
7.50%
4%
19%
15%
20%
22.50%
10%
24%
8%
16%
4.20%
2.20%
2007
2017
2015
2010
2019
2022
2020
2013
1999
2000
2019
2009
2015
2015
2015
2020
2011
2013
2020
2019
2015
2011
No
No
Yes
Yes
Yes
Yes
No
No
No
No
Yes
Yes
No
Yes
Yes
Yes
Yes
No
Yes
Yes
Yes
Yes
Percent of 2004 U.S. Retail Sales in States with Renewable Energy Standards
All States with Renewable Energy Standards:
States with Credit-trading Policies:
48.8%
25.5%
Sources and Notes:
*Illinois has established a renewables requirement with no specific
enforcement measures, but utility regulatory intent and authority appears sufficient.
**Minnesota has established both a requirement and a goal.
For a list of states with RPS policies, see Union of Concerned Scientists,
"State Minimum Renewable Energy Requirements (as of April 2006)".
2004 retail sales data is from Energy Velocity.
48
In 14 of the states (comprising about 25 percent of U.S. retail load), the RPS includes tradable renewable energy credits
(RECs), which can ease compliance. An REC represents one MWh generated from an eligible source, and can be
decoupled from the actual generation and sold separately for compliance. A retail utility can use any combination of
renewable power actually purchased and RECs; likewise a renewable generator can sell RECs separately from its power
sales. Tradable RECs can simplify transactions and lower costs by creating, in effect, a separate wholesale market for the
renewable attributes of eligible generation.
47
Chapter 4: Generation Investment
Estimates vary on the amount of renewable electric generating capacity that will be developed to attain the
standards and goals already promulgated. According to EIA’s Annual Energy Outlook 2006, about 10 GW
of additional renewable capacity is likely over the next 20 years, while the Union of Concerned Scientists
estimates about 30 GW of new renewable capacity over the next 10 years.49
It is too early to tell just how RPS policies will contribute to electricity price increases, since many of the
ambitious targets lie in the future. Similarly, experience in renewable energy credit (REC) markets and
resultant price dynamics is limited to those few states with active REC markets. However, in the majority of
cases, renewables (or equivalently, RECs) will be purchased at prices above the wholesale cost of
conventional generation and thus will increase the overall cost of serving load in states where such policies
have been enacted. The additional expenditures from mandatory renewable obligations represent additional
costs that should be recovered in rates.
Green Electricity Marketing
Many utilities are also offering new products in the form of “green” electricity options whereby they fill all
or part of the customers’ load with renewable supply (or RECs) and charge slightly higher rates to reflect the
higher costs of renewable power. These programs have grown rapidly with both residential and business
customers, and in some regions the renewable rates have actually proven to be quite competitive as recent
fuel price increases have been reflected in standard customer tariffs. While the higher rates paid by
consumers reflect voluntary preferences, these programs are helping to increase renewable market share in
some states.
On-Site Customer Generation
The need for additional utility generation and transmission will be mitigated to some extent by increased
development of small, onsite customer generation. Such generation is typically known as distributed
generation (DG). Examples of DG include microturbines, biomass-based generators, small wind turbines,
solar thermal electric devices, and backup generators at office buildings, industries, and hospitals. In
contrast to large, central-station power plants, distributed power systems typically range from less than a
kilowatt to tens of megawatts in size. EIA projects that 5.5 GW of DG, or slightly less than two percent of
all new generating capacity, will be installed over the next 25 years.50
In addition to reducing the need for generation investment, optimally sited DG can reduce the need for
transmission and distribution investment while resolving some system constraints and reducing line losses.
The current efficiency of microturbines in the range of two to 75 kW is rather low but as their efficiency
improves these small generators will become more attractive alternatives to grid-based electricity services.
49
Energy Information Administration, Annual Energy Outlook 2006, and estimates on Union of Concerned Scientists’ Web
site at ucsusa.org. The discrepancies between the estimates primarily reflect differences in assumed impacts of RPS
policies, different technology mixes arising to satisfy the RPS generation requirements, and differences in measuring the
existing renewable capacity base.
50
Energy Information Administration, Annual Energy Outlook 2006, February 2006, Table A9.
48
Why Are Electricity Prices Increasing? An Industry-Wide Perspective
In addition, public policy objectives (e.g., development of small-scale renewable generators) are likely to
foster the development of DG.
Section 1251 of EPAct 2005 encourages the development of small, onsite generation by requiring states to
consider if utilities should make net metering services available upon request to any customer. Net metering
allows electric customers to sell to a local utility (or electricity supplier) any excess electricity generated by
an onsite generation source. Excess electricity produced by the onsite generator will spin the customer’s
meter “backwards” such that the customer is a net seller of electricity to the local utility at such times. Net
metering is a policy that many states already have implemented to encourage the use of small renewable
energy systems. Approximately 40 states have adopted some form of net metering law for small wind and/or
photovoltaic technologies whereby the customer receives a credit for excess power sold to the utility.51
Under most state rules, all retail customers are eligible for net metering; however, some states restrict
eligibility to particular customer classes. Customer participation in net metering programs has grown
significantly. In 2004, a total of 15,286 customers were in net metering programs—a 132-percent increase
from 2003. Residential customers accounted for 89 percent of all customers participating in such
programs.52
Net metering offers onsite generators a convenient way to account for energy production, allowing excess
energy produced to be offset against energy purchases made at other times. Net metering also can be an
inexpensive way to sell excess energy in quantities that are too small or intermittent to market directly. For
these reasons, net metering can promote the development of small-scale renewable technologies that can
defer or displace a modest amount of central-station generation and transmission capacity.
The use of net metering with current metering technology is problematic, however, because today’s meters
cannot account for the difference between high-cost peak and low-cost off-peak electricity, nor can they
account for the difference in wholesale and retail electricity costs. For example, a conventional meter only
can record that over a given month an onsite generator sold a net of 100 kWh to the local utility, but will
have no record of when the 100 kWh was sold. Sales at 4 p.m. on a hot summer weekday will have a much
higher value than sales at 3 a.m. on a Saturday morning. With conventional metering, an onsite generator
will have to be compensated at an average wholesale (or retail) rate, which will not accurately reflect the
value of the energy provided by the generator. Thus, another benefit of advanced metering technology
discussed in Chapter 6 is that it will enable more accurate valuation and compensation of energy provided by
onsite generators.
51
52
See www.dsireusa.org.
Energy Information Administration, Green Pricing and Net Metering Programs 2004, March 2006.
49
CHAPTER 5
Transmission Investment
Overview of the Transmission Grid
Consumers depend on the high-voltage transmission grid for access to reliable and reasonably priced
supplies of electricity. The Northeast blackout of August 14, 2003, which was caused by operational failures
rather than inadequate infrastructure, disrupted service to more than 50 million customers over an area
extending from Michigan to western Massachusetts, including Detroit, Toronto, Cleveland, Ottawa, Buffalo,
and New York City, with costs estimated to be between $4 billion and $10 billion.53 This blackout
demonstrated the severe costs that wide-scale transmission disruptions can entail.
The U.S. and Canadian electric transmission grid includes more than 200,000 miles of high-voltage (230 kV
and greater) transmission lines that ultimately serve more than 300 million consumers.54 This system was
built over the past 100 years, primarily by vertically integrated utilities that generated and transmitted
electricity locally for the benefit of their native load customers.55 Today, 134 control areas or balancing
authorities56 manage electricity operations for local areas and coordinate reliability through the eight regional
reliability councils of NERC.
Interconnections between neighboring utilities have long existed, but were initially created to increase
reliability and allow utilities to share excess generation through infrequent economy transactions. Over the
past 15 years, successive federal policy initiatives have promoted the development of regional power
markets. The transmission system is a critical facilitator of these power markets as well as a means of
delivering power reliably to retail customers.57
53
U.S.-Canada Power System Outage Task Force, Final Report of the August 14, 2003 Blackout in the United States and
Canada: Causes and Recommendations, April 2004, p. 1.
54
Clark W. Gellings and Kurt E. Yeager, “Transforming the Electric Infrastructure,” Physics Today, December 2004,
pp. 45-46.
55
“Native load” customers are those customers whom the utility is obligated to serve either by statute or by contract.
56
A balancing authority, formerly known as a control area, is an electric system or systems, bounded by interconnection
metering and telemetry, capable of controlling generation to maintain its interchange schedule with other control areas and
contributing to frequency regulation of the interconnection.
57
U.S. Department of Energy, National Transmission Grid Study, May 2002, Executive Summary.
51
Chapter 5: Transmission Investment
The U.S. electricity delivery system, which consists of the transmission grid and the downstream distribution
system, is a $360-billion asset.58 Unfortunately, this power delivery system is characterized by an aging
infrastructure and largely reflects technology developed in the 1950s or earlier. According to DOE, 70
percent of transmission lines are 25 years or older, 70 percent of power transformers are 25 years or older,
and 60 percent of circuit breakers are more than 30 years old.59 The strain on this aging system is beginning
to show, particularly as market participants and regulators ask it to perform functions (e.g., facilitate
competitive regional power markets) for which it was not originally designed.
Transmission Investment Trends and Drivers
Transmission investment declined steadily for approximately 25 years, increasing only over the last few
years.60 Between 1975 and 1999, nominal investment for investor-owned utilities (IOUs) fell at an average
rate of $83 million per year. The trend reversed itself from 1999 to 2003 as nominal transmission investment
increased by an average of $286 million per year and totaled nearly $18 billion over this period.61 Figure 5-1
illustrates that transmission mileage has not dramatically increased in recent years, relative to growth in load.
“Normalized” transmission capacity, or the number of transmission line miles per unit of demand, declined
by almost 19 percent between 1992 and 2002.62
Figure 5-1
Transmission Mileage and Demand
1,000
200
190
NERC ES&D
Transmission Line
Miles
850
180
170
800
160
750
150
700
140
650
130
NERC ES&D
Peak Demand
600
Historical
120
Projected
2013
2011
2009
2007
2005
2003
2001
1999
1997
100
1995
500
1993
110
1989
550
1991
Demand (GW)
900
Transmission Mileage (000s)
950
Year
Source and Note: NERC Electricity Supply & Demand 2004. Circles reflect forecast values.
58
Gellings and Yeager, p. 46.
Center for Smart Energy, The Emerging Smart Grid: Investment and Entrepreneurial Potential in the Electric Power Grid
of the Future, October 2005, p. 9.
60
See Peter Fox-Penner, “Rethinking the Grid,” Electricity Journal, March 2005.
61
Eric Hirst, U.S. Transmission Capacity: Present Status and Future Prospects, August 2004, p. 7.
62
Hirst, August 2004, p. 9.
59
52
Why Are Electricity Prices Increasing? An Industry-Wide Perspective
While NERC believes that the existing transmission grid is sufficient to provide reliable service in the near
term, the Council acknowledges that some portions of the grid will not be able to support all desired market
transactions.63 In addition, NERC believes that regional transmission networks will be operated near or at
their limits more frequently in the foreseeable future.64 This implies that the transmission system—while
currently reliable—will experience greater congestion.65 Public data on congestion costs generally are
available only in regions with centralized, RTO-administered energy markets. As illustrated in Figure 5-2,
congestion costs in the RTO markets are significant and have increased over time. The figure displays
reported congestion costs for ISO New England, the New York ISO, PJM, and the California ISO for all
years for which data are available from 2001 to 2005. Notice that total congestion costs are nearly $1 billion
per year in New York and more than $2 billion per year in PJM. Although we do not have comparable data
for other parts of the United States not shown in this figure, there are indications that congestion is increasing
everywhere on the North American power grid.66
Figure 5-2
Annual Congestion Costs/MWh of Load by RTO/ISO
Congestion Cost / MWh of Load
$5.00
$4.50
$4.00
$3.50
$3.00
$2.50
$2.00
$1.50
$1.00
$0.50
Notes and Sources:
Respective ISO websites.
2005 data for NYISO not yet available.
ISO-NE
NYISO
PJM
2005
2004
2003
2002
2001
$0.00
CAISO
63
North American Electric Reliability Council, 2005 Long-Term Reliability Assessment, September 2005, p. 6.
Id., p. 5.
65
Transmission congestion occurs when the power grid cannot accommodate all desired transactions between power buyers
and sellers (or when a vertically integrated utility cannot move all of its low-cost generation to its customers). When
congestion occurs, system operators must “redispatch” generation—i.e., use relatively high-cost generation in place of
lower-cost generation—to serve total customer demand within the limitations of the transmission system. The incremental
cost associated with redispatching generation is the cost of congestion.
66
Evidence that transmission congestion is increasing in regions without RTOs is provided by the steady increase in the
number of transmission loading relief (TLR) procedures called by Security Coordinators over the last eight years. TLRs
are called when the transmission system cannot simultaneously accommodate all desired transactions.
64
53
Chapter 5: Transmission Investment
Some degree of congestion cost is efficient, to the extent that the cost of alleviating all congestion on the grid
may be prohibitive. Such congestion costs, properly measured, can provide an important price signal for
additional generation or transmission investment in specific areas. Nevertheless, increasing congestion and
the aging power delivery infrastructure have spurred calls within the industry and by many federal and state
agencies to expand transmission investment.67
Transmission Investment Looking Forward
In response to these conditions, utilities are expanding their transmission investments substantially. In a
recent survey (May 2005), the Edison Electric Institute (EEI) shows that IOUs have spent or plan to spend
$29 billion in transmission infrastructure from 2004 to 2008, a 60-percent increase over the previous five
years.68 Figure 5-3 depicts the historical investment trends in both real and nominal terms, along with EEI’s
forecasts based on its survey of IOUs. This figure highlights that the recent upturn in transmission
investment coincided with the surge in generation, and that high levels of investment are expected to
continue.
Figure 5-3
Construction Expenditures for Transmission
By Investor-Owned Electric Utilities
$8
$7
$6
$ (Billions)
Real (2004$)
$5
EEI
Projections
$4
$3
$2
$1
Nominal
Source: EEI.
2007
2005
2003
2001
1999
1997
1995
1993
1991
1989
1987
1985
1983
1981
1979
1977
1975
$0
Year
Net book value of investor-owned transmission assets totaled approximately $43 billion in 2003. Thus,
planned investment over the 2004 to 2008 period is 62 percent of year 2003 net book value. Both stand-
67
U.S. Department of Energy, Office of Electric Transmission and Distribution, National Electric Delivery Technologies
Roadmap: Transforming the Grid to Revolutionize Electric Power in North America, January 2004, p.3.
68
Edison Electric Institute, EEI Survey of Transmission Investment: Historical and Planned Capital Expenditures (19992008), May 2005.
54
Why Are Electricity Prices Increasing? An Industry-Wide Perspective
alone transmission companies and vertically integrated utilities are planning significant growth in
investment. EEI survey respondents indicated that, on average, only a small portion of this total planned
transmission investment, 6.5 percent, is attributed to direct generator interconnections. This indicates that
the bulk of projected investments in the nation’s transmission infrastructure will support the integration of
new generator additions through network upgrades, improved transfer capability between regions, improved
grid reliability, and enhanced local, regional, and inter-regional markets.
Recent transmission plans prepared by the RTOs provide further evidence of the ambitious plans underway
to expand and reinforce regional power networks. The Midwest ISO has identified almost $3 billion of
planned or proposed investments through 2009, primarily to maintain reliability.69 Other regions, notably the
Northeast, also have aggressive plans to build or upgrade both transmission and distribution. For example,
PJM recently completed its 2005 plan to meet reliability needs through 2010, by approving a total of $1.8
billion of transmission upgrades in its region.70 ISO New England’s most recent transmission plan identifies
272 needed transmission projects with a total cost of about $3 billion.71 As part of the plan to import more
power into heavily populated Southern California, the California ISO recently approved a major expansion
of the Palo Verde-Devers transmission line for a cost of $680 million.72
This investment will yield a substantial amount of new transmission capacity. According to NERC, more
than 7,122 miles of new transmission (230 kV and above) are proposed to be added through 2009, with a
total of about 12,484 miles added over the 2005 to 2014 time frame. This represents a 5.9-percent increase
in the total miles of installed extra-high-voltage transmission lines in North America over the 2005 to 2014
period.73 Nearly 1,200 miles of new or upgraded transmission lines will be added in 2006 alone.74 Some of
the investment cited above will be dedicated to other means of enhancing transmission capacity, such as
upgrading or rewiring existing lines and replacing transformers.
Factors Driving Increased Transmission Investment
Several factors are contributing to the recent and expected future increase in transmission investment. These
factors include: (a) the regionalization of transmission planning and investment; (b) the return to larger and
more remote baseload generation sources; and (c) new transmission policies and incentives at the federal
level.
With respect to the first factor, transmission planning is evolving in important ways that will tend to place
more emphasis on identifying the transmission upgrades needed to enhance regional trade and reduce
congestion. Traditionally, transmission planning was performed by vertically integrated utilities, which built
69
Midwest Independent Transmission System Operator, Inc., Midwest ISO Transmission Expansion Plan 2005, June 2005.
http://www.pjm.com/contributions/news-releases/2006/20060407-pjm-authorizes-one-year-total-of-1.7-billion-i.pdf.
71
ISO New England, 2005 Regional System Plan, Executive Summary.
72
http://www.caiso.com/docs/2005/02/25/200502251524204169.pdf.
73
North American Electric Reliability Council, 2005 Long-Term Reliability Assessment, September 2005, p. 6.
74
North American Electric Reliability Council, 2006 Summer Assessment, May 2006, p. 3.
70
55
Chapter 5: Transmission Investment
the transmission capacity needed to deliver power from local generating plants to their native load customers.
Most recent RTO plans also have focused primarily on identifying the transmission upgrades needed to
maintain reliability or interconnect new generators to the regional network, as the RTO investment numbers
cited herein largely are for reliability-driven investments. This is changing. With some prodding by FERC,
RTOs are expanding beyond traditional, reliability-based planning models and studies to explicitly include
economic considerations in their transmission plans. For example, PJM, the largest U.S. RTO, has expanded
its planning process to include an analysis of economic upgrades—meaning upgrades not needed to maintain
adherence with PJM’s reliability criteria but those that may help to reduce electricity supply costs to
customers.
In addition, utilities are aggressively pursuing opportunities to reduce intra- and inter-regional bottlenecks.
One prominent example is the 550-mile, 765-kV line proposed by American Electric Power (AEP), which
would run from West Virginia to New Jersey. The line would cost approximately $3 billion and would
increase Midwest-to-East transfer capability by approximately 5,000 MW, thereby allowing more low-cost,
coal-fired power to reach eastern PJM, which tends to have relatively high energy prices. AEP plans to have
the line in service by 2014. Another example is the 500-kV line proposed by Allegheny Energy, which
would span 330 miles, all within Allegheny’s service territory, from West Virginia to central Maryland.
This line is projected to cost $1.4 billion, with the first segment in place by 2013. A further example is the
230- mile, 500-kV line proposed by Pepco Holdings (PHI), which would run from Northern Virginia, cross
Maryland, and travel up the Delmarva Peninsula to New Jersey. PHI claims that the line, which is estimated
to cost $1.2 billion, would significantly increase reliability in the eastern mid-Atlantic region and would
complement proposals from AEP and Allegheny to improve West-to-East transfer capability in PJM. If
approved, the line could be built in stages beginning in 2008.
The second factor spurring a demand for long-distance lines is the shift away from gas-fired generation to
large, baseload coal-fired and nuclear generation and renewable generation. Over the last 15 years, most of
the new generation capacity added in the United States has been gas-fired capacity. Today there is much
more interest in building coal-fired capacity, and such capacity comprises a far more significant share of new
generating capacity in development or under construction than in the recent past. Most of this new coal-fired
capacity will be distant from population centers for environmental and/or fuel supply reasons This will
require additional long-distance transmission capacity. Similarly, wind farms are located at remote, sitespecific resources. Thus, the increase in natural gas prices is driving the mix of new generating capacity to
resources that are likely to require a significant amount of new network transmission capacity to deliver their
output to load centers.
The incremental cost associated with this transmission capacity appears significant. As an example, the
Western Governors Association concluded that a generation expansion plan in the western United States
featuring coal, wind, and geothermal generation would require approximately $8 billion to $12 billion in
transmission investment over the next 10 years, whereas a generation expansion plan featuring gas-fired
generation would require only about $2 billion of transmission investment.75
75
Western Governors Association, Conceptual Plans for Electricity Transmission in the West, August 2001, p. 4.
56
Why Are Electricity Prices Increasing? An Industry-Wide Perspective
Policy Initiatives to Facilitate Transmission Investment
EPAct 2005 included several provisions to facilitate the siting and construction of new transmission
facilities. Prior to EPAct 2005, state and local agencies had exclusive jurisdiction over transmission siting,
with the exception of lines that crossed federal lands or international boundaries. Section 1221 of EPAct
2005, however, gives FERC the authority to site new transmission lines in congested areas or regions
designated by DOE as corridors of national interest. By August 2006 (and every three years thereafter),
DOE must prepare a report identifying congested lines or corridors. The legislation appears to give DOE
wide latitude to designate any area experiencing congestion as a corridor of national interest. FERC can
exercise its siting authority if a proposed line that would relieve congestion in a national corridor of interest
does not receive the necessary approvals from state and local authorities within one year of filing the
necessary applications. DOE also is given considerable authority to expedite and coordinate the siting of
transmission facilities over federal lands.
EPAct 2005 also provides for three or more contiguous states to form an interstate compact for the purpose
of establishing a regional transmission siting agency. States that enter such compacts, which must be
approved by Congress, are exempt from FERC’s backstop siting authority. Such regional siting agencies
must have the authority to issue permits necessary for the siting of transmission facilities (i.e., the regional
agency acts on behalf of the represented states).
Section 1241 of EPAct 2005 directs FERC to establish incentive-based rate treatments for transmission
investment and deployment of new transmission technologies. FERC has issued a proposed rule that
establishes a menu of potential incentives that would be available for new investments on a case-by-case
basis, such as 100-percent recovery in rate base of prudently incurred Construction Work in Progress and
accelerated recovery of depreciation expenses.
The goal of EPAct 2005’s siting and incentive ratemaking provisions is to reduce the perceived regulatory
barriers to the construction of new transmission capacity. Much uncertainty remains as to the impact of the
financial incentives and FERC’s new siting authority, especially given the lack of precedent for the latter.
Thus, time will be needed to determine the effectiveness of these provisions. But the AEP, Allegheny, and
PHI announcements demonstrate that the industry is willing to invest significant dollars in new lines
designed to expand regional trade and markets for low-cost generation resources.
Transmission Grid and Retail Rates
Evidence suggests that the transmission-related components of retail rates are small, but are growing rapidly.
The costs of the grid are reflected in retail rates through return on transmission rate base, as well as through
O&M expenses attributed to transmission services. Since 1998, EIA has published estimates of nationwide
retail rates, broken down by service category. In 2004, for example, retail rates were divided into service
categories as follows:
ƒ Generation:
ƒ Transmission:
ƒ Distribution:
4.97 cents/kWh
0.54 cents/kWh
2.07 cents/kWh
57
Chapter 5: Transmission Investment
EIA also provides short-term and long-term projections of these service category components. As
illustrated in Figure 5-4, the portion of retail rates attributed to transmission is expected to increase rapidly.
Indeed, the cumulative increase in the transmission portion of retail rates is expected to be approximately 42
percent from 2004 to 2014. This increase far exceeds the 17-percent increase expected in the generation
portion and distribution portion of retail rates from 2004 to 2014. (EIA projects a 19-percent increase in
retail rates over this period.) The sharp increase in the transmission portion of retail rates reflects both the
significant increase in transmission investment, which comes in response to the surge in generation capacity
during the late 2000s, and the fact that the value of existing transmission assets is smaller than the asset
value for either generation or distribution.
Figure 5-4
Change in Nominal Retail Rates by Service Category
1.5
Historical
1.4
Forecast
Index (1998 = 100)
1.3
Generation
1.2
Distribution
1.1
1
Transmission
0.9
0.8
0.7
0.6
2014
2013
2012
2011
2010
2009
2008
2007
2006
2005
2004
2003
2002
2001
2000
1999
1998
0.5
Source: EIA.
Transmission Grid of the Future
The preceding discussion referred to the transmission investment needed over the next five to 10 years to
achieve three primary objectives: (1) maintain reliable service; (2) interconnect new generators to the grid,
including large baseload generators and remotely sited power plants; and (3) reduce congestion and foster
economical wholesale power trade in regional power markets. In the long-run, however, the technology of
the power grid itself must evolve to meet the needs of our digital, information-driven economy. The
knowledge-based economy of the future increasingly will require a more technology-driven delivery system
that links information technology with energy delivery. The revolution in information technologies that has
transformed other industries has yet to occur fully in the electric power business.
As an example, some of today’s transmission system still relies on electro-mechanical switches—the same
switches that were eliminated from consumer television sets 20 years ago. The digital controls that will
replace these will become the foundation of a new “self-healing” power delivery system that will enable
58
Why Are Electricity Prices Increasing? An Industry-Wide Perspective
innovative technologies and processes to flourish throughout the U.S. economy. Moreover, this technology
will address the combined reliability, capacity, security, and other vulnerabilities of today’s power delivery
systems.76
The concept of the smart-power delivery system includes automated capabilities to recognize problems, find
solutions, and optimize the performance of the power delivery system in real time. The basic building blocks
include advanced sensors, data-processing and pattern-recognition software, and solid-state power flow
controllers to reduce congestion, react in real time to disturbances, and redirect the flow of power as needed.
These new technologies will enable system operators to (1) optimize the overall performance and resilience
of the system; (2) instantly respond to disturbances to minimize their impact; and (3) restore the system after
a disturbance.77 With greater real-time information, system operators will be able to sense, predict, diagnose,
and mitigate issues that might previously have caused an outage or blackout.78 Table 5-1 provides an
overview of the key differences between today’s power grid and the smart grid of the future.
Table 5-1: The “Smart Grid” of the Future
20th Century Grid
21st Century Grid
Electromechanical
One-way communications (if any)
Built for centralized generation
Radial topology
Few sensors
“Blind”
Manual restoration
Somewhat prone to failures and blackouts
Check equipment manually
Limited control over power flows
Limited price information
Few customer choices
Digital
Two-way communications
Accommodates distributed generation
Network topology
Monitors and sensors throughout
Self-monitoring
Semi-automated restoration and, eventually, self-healing
Adaptive protection and islanding
Monitor equipment remotely
Pervasive control systems
Full price information
Many customer choices
Source: Center for Smart Energy, The Emerging Smart Grid: Investment and Entrepreneurial Potential in the Electric Power Grid of the
Future, October 2005, p. 2.
76
Electric Power Research Institute, Electricity Sector Framework for the Future: Volume 1, Achieving a 21st Century
Transformation, August 6, 2003, p. 28.
77
Id., p. 30.
78
Center for Smart Energy, The Emerging Smart Grid: Investment and Entrepreneurial Potential in the Electric Power Grid
of the Future, October 2005, pp. 11-12.
59
Chapter 5: Transmission Investment
A variety of new technologies, some of which are in the initial commercialization stage, will facilitate the
transition from today’s generally reliable but somewhat antiquated power delivery system to tomorrow’s
“Smart Grid.” Some of these technologies are discussed in the text box below. Generally speaking, these
technologies: (1) increase system throughput or otherwise allow better utilization of existing transmission
facilities; (2) allow operators to better monitor system conditions; or (3) enable the grid to recover more
quickly from disturbances. While it is unclear how quickly we will transition to a “Smart Grid,” what is
apparent is that the cost of doing so will be very significant. A study performed by the Electric Power
Research Institute (EPRI) suggests that research, development, and deployment costs to transform the
transmission system into the “Smart Grid” of the future would approach $200 billion over a 20-year period.79
Technologies That Enhance Increased System Throughput
Some emerging transmission technologies will enable the existing system to carry more electricity in
a reliable manner and thus should ease some of the current and growing stresses on the bulk power
grid. One promising group of technologies—Flexible AC Transmission Systems (FACTS)—are in
the initial commercialization stage, but more research is needed to reduce their costs before they
achieve wide penetration in the marketplace. FACTS devices are a family of solid state power
control devices that provide enhanced power control capabilities to high-voltage AC grid operators.
FACTS provide nearly instantaneous control of AC power flows, far faster than traditional, electromechanical AC switches. By providing transmission operators with quicker response capability,
FACTS enables them to operate the system closer to otherwise applicable limits (e.g., thermal
limits), effectively getting more transmission capability out of existing power lines. In addition,
FACTS devices can increase the transfer capability of existing lines by a modest amount (up to 10
percent) as well as enhancing stability and overall reliability.1
Some of the new FACTS devices could reduce the need for keeping uneconomical “reliability must
run” (RMR) generating units in service in transmission-constrained areas. One such technology is
the D-VAR® Systems (Dynamic Volt Ampere Reactive), a new, modular FACTS device that is
replacing Static VAR Compensators. These devices inject leading or lagging voltage precisely
where it is needed in a grid. D-VAR systems can be packaged in mobile trailers or installed
permanently in substations. Several dozen D-VAR systems are now in use in the United States,
Great Britain and Canada to enhance power transfers into, across, and out of congested areas while
improving grid reliability and power quality. The D-VAR solution can be used to reduce or
eliminate the operation of costly RMR units.2
Similarly, the Super VAR® dynamic synchronous condenser is a new application that helps to
stabilize grid voltage, increase service reliability and maximize transmission capacity by acting as a
“shock absorber” for grid voltage fluctuations. The first Super VAR system prototype is undergoing
evaluation on TVA’s grid. This tool is able to supply a large amount of reactive power support very
79
Electric Power Research Institute, Electricity Technology Roadmap: Meeting the Critical Challenges of the 21st Century,
2003 Summary and Synthesis, pp. 1-7.
60
Why Are Electricity Prices Increasing? An Industry-Wide Perspective
efficiently and can operate at several times its nominal rating for short periods to dampen out more
severe transient disturbances.3
These dynamic VAR technologies offer new ways to reduce the risk of forecast uncertainty.
Modular and compact, they are easily studied, sited, and installed within a planning cycle. These
assets can be relocated as system needs change. These technologies, in short, offer a flexible, “just
in time” approach to grid planning and can help grid operators maintain reliability in real-time while
planners defer costly and irreversible investments to major grid resources.
Other promising new technologies that will enable increased system throughput, but which are
farther removed from commercial deployment than the FACTS technologies, include:
ƒ High Temperature Superconducting Cables (occupy less space; reduced risk of damage to the
environment);
ƒ High Ampacity Conductors (reduce sag, permit greater load of lines; longer service life);
ƒ Dynamic Line Rating (use real-time information, allowing higher thermal capacity of
transmission lines and substation equipment);
ƒ Video Sag Monitoring (extends effectiveness of DLR);
ƒ Solid State Superconducting Fault Current Limiter (limit fault current contributed by new
generation; add performance beyond that of conventional breakers); and
ƒ Solid State Power Electronics Circuit Breaker (reduce response time to faults; lower
maintenance costs and improved reliability).4
One of the more significant breakthroughs in advanced materials for electric power is the emergence
of high temperature superconductors. These superconducting materials can replace existing grid
segments with greatly enhanced capabilities, thereby giving the grid more flexibility, reliability, and
efficiency, which would mean less electricity losses and less primary energy use, thus lowering the
environmental impacts of power production.
Technologies That Allow Operation Closer to System Limits
Another group of advanced technologies will enable operators to run the transmission system closer
to its limits by reducing the conservative assumptions or margins used to set existing limits,
allowing these limits to be increased and thereby expanding the usable capacity of the transmission
system. One set of technologies focuses on accurate monitoring to improve engineering
management of the transmission system. These technologies will detect abnormal system conditions
and will indicate when security limits are being reached in time. They include Wide Area
Measurement Systems (WAMS) and Topology Estimators. WAMS, which was initially developed
by BPA, is a system based on high-speed monitoring of a set of measurement points. WAMS
detects abnormal conditions as they arise and thereby provides a strong foundation on which to build
the real-time wide-area monitoring system for the self-healing grid.5
A second set of technologies goes a step further, enabling operators to use real-time engineering
information to assess economic conditions, including congestion, to support competitive wholesale
market operations. These technologies range from integrated engineering and economic methods for
power system operation, to visualization and communications tools, to virtual RTO technology and
61
Chapter 5: Transmission Investment
market simulation. For example, monitoring systems are under development that would enable
system operators to dynamically determine line and transformer capacity. In addition, powerelectronics controllers, based on solid-state components, will yield control of the power delivery
system with the speed and accuracy of a microprocessor.
Technologies That Reduce Load at Critical Times
In addition to customer-based demand-response programs, such as programs that give customers
financial incentives to reduce their demand at times of high system-wide demand, there also are
emerging energy storage technologies that system operators could use to reduce demand. One such
technology is super-conducting magnetic energy storage (SMES). This device stores power taken
from the grid in a super-conducting coil and injects it back into the grid when needed (e.g., when
voltage sags). It thereby provides additional support to the grid and can effectively be used to
expand transmission capacity. The initial deployment of a group of six SMES units in Wisconsin
increased transmission capacity by approximately 15 percent.6 Other load reduction technologies
include compressed-air storage and flywheels.
1
2
3
4
5
6
62
Philip M. Marston, Esq., Of Chips, Hits, Bits and Bytes: Building the Powerline Paradigm, Report and
Recommendations of the Grid Enhancement Forum of the Center for the Advancement of Energy
Markets, Draft for Discussion, June 24, 2002, pp. 24-25. (Hereafter, CAEM Report.)
John B. Howe, Using Dynamic VAR Technologies to Boost Grid Reliability, Utility Automation and
Engineering/T&D, May-June 2005.
Id.
The Consumer Energy Council Transmission Infrastructure Forum, Keeping the Power Flowing:
Ensuring a Strong Transmission System to Support Consumer Needs for Cost-Effectiveness, Security
and Reliability, January 2005, pp. 90-91.
Electric Power Research Institute, Electricity Technology Roadmap, 1999 Summary and Synthesis.
CAEM Report, p. 40.
CHAPTER 6
Distribution Investment
Trends in Distribution System Investment
The transmission system delivers power from generators to local distribution systems, which in turn deliver
power to residential, commercial, and industrial customers. Specifically, the transmission system feeds
substation transformers that reduce voltage and spread the power from each transmission line to many
successively smaller distribution lines. The distribution system usually is considered to begin where voltage
is reduced to 37 kV, but the important distinction is that distribution involves delivering the power to retail
customers, while transmission involves moving bulk power to distribution systems. The distribution system
also includes metering, billing, and other related infrastructure and software associated with retail sales and
customer care functions.
Continual investment in distribution facilities is needed, first and foremost, to keep pace with growth in
customer demand. Figure 6-1 shows the pattern of investment in distribution assets over the last 30
years. In real terms, investment began to increase in the mid-1990s, preceding the corresponding boom
in generation. This steady climb in investment in distribution assets shows no sign of diminishing. The
need to replace an aging infrastructure, coupled with increased population growth and demand for
power quality and customer service, is continuing to motivate utilities to improve their ultimate delivery
system to consumers.
Figure 6-1
Construction Expenditures for Distribution
By Investor-Owned Electric Utilities
$16
Nominal Dollars
$14
Real Dollars (2004)
$10
$8
$6
$4
$2
Source: EEI.
2005
2003
2001
1999
1997
1995
1993
1991
1989
1987
1985
1983
1981
1979
1977
$0
1975
$ (Billions)
$12
Year
63
Chapter 6: Distribution Investment
As shown in Chapter 3, continued load growth will require continued expansion in distribution system
capacity. If recent investment trends persist, distribution investment will average $14 billion per year over
the next 10 years. This is almost triple the projected amount of annual investment in new transmission
capacity and is likely to exceed capital spending on generation capacity over the next decade as well. This
level of distribution investment would lead to a cumulative 3.5-percent increase in retail rates over the next
10 years.
Other factors apart from load growth, such as aesthetics, storm damage, and local land use, will spur
spending on distribution infrastructure. In some cases, utilities are being directed to place new and/or
existing distribution lines underground, particularly in urban areas. Placing existing power lines
underground is expensive, costing approximately $1 million per mile—a five- to ten-fold increase over the
cost of a new overhead power line.80 Moreover, at a cost of $1 million per mile, a new underground system
would require an investment of more than 10 times what the typical U.S. IOU currently has invested in
distribution plants and would compel the utility to increase its rates.81
Need to Modernize Distribution Systems
Distribution investment also will be needed to meet the greater demand for increased reliability. The impact
of power disturbances on customers has grown steadily over time due to the increased use of digital
technology. The current U.S. electricity infrastructure was designed to serve analog, or continuously
varying, electric loads, and does not consistently provide the level of digital-quality power required by the
nation’s digital manufacturing assembly lines, information systems, and, increasingly, home appliances.82
Digital devices are highly sensitive to even the slightest interruption of power; an outage of less than a
fraction of a single cycle can disrupt their performance. They also are quite sensitive to variations in power
quality. Digital quality power has the same overall voltage as today’s power and is indistinguishable from
analog appliances, but has reduced levels of signal variations that adversely affect digital circuits. An
enhanced power system capable of delivering this higher quality power will stimulate faster and more
widespread use of productivity-enhancing digital technology.
It is not an exaggeration to say that we are experiencing a “digitalization of society”—today, there are more
than 12 billion microprocessors in the United States alone.83 For every microprocessor inside a computer, 30
operate in standalone applications. Digital-quality power now represents about 10 percent of total electric
load in the United States. EPRI projects that digital-quality power load will reach 30 percent in 2020 under
business-as-usual conditions.84
80
Brad Johnson, Out of Sight, Out of Mind? A Study on the Costs and Benefits of Undergrounding Overhead Power Lines,
Prepared for the Edison Electric Institute, January 2004, p. 14.
81
Id., p. 14.
82
Gellings and Yeager, p. 50.
83
Gellings and Yeager, p. 49.
84
Gellings and Yeager, p. 49.
64
Why Are Electricity Prices Increasing? An Industry-Wide Perspective
Consistent with this trend, a recent study commissioned by DOE found that residential energy use on
information technology (IT) applications increased substantially in the last few years, reflecting the dramatic
increase in the use of personal computers and related devices. The study estimated that home IT equipment
consumed about 42 terawatt hours (TWh) of electricity in 2005, compared to 16.5 TWh in 2001.85 That is,
residential IT equipment accounted for about three percent of residential electricity consumption and one
percent of U.S. electricity consumption in 2005. The study projects that, by 2010, residential IT energy
consumption could rise to 101 TWh, under a scenario which assumes widespread high-bandwidth
connectivity that enables effective exchange of large quantities of data and programs run on desktop
computers.
Three new technologies already under development will enable utilities to provide digital-quality power and
other enhanced distribution value. One is distribution automation. Distribution automation uses advanced
sensors and control software to improve power suppliers’ ability to detect and correct disturbances more
quickly, thus reducing customer outages and power quality problems. These capabilities lead to rapid
disturbance isolation and restoration capabilities.
The second technology development path is custom power, a family of power electronic controllers designed
for service on distribution systems. These devices and systems can provide real-time network control,
protect sensitive customer equipment from network disturbances, and protect distribution feeders from power
disturbances arising on the customer’s premises. Custom power systems improve power quality for
customers with special needs—for example, an industrial park with high technology companies.
The third path is the development of generation and storage technologies for distributed applications. These
devices will move the power supply closer to the point of use, enabling improved power quality and
reliability, and providing the flexibility to meet a wide variety of customer and distribution system needs.86
This path is one of the main drivers of distributed generation, which is discussed further in Chapter 4.
Distribution systems will need to be updated to seamlessly integrate an array of locally installed, distributed
power generation (such as fuel cells and renewables) as power system assets. In some cases, utilities will
make the investments in these new technologies through their own programs or subsidiaries; in others,
customers will invest in these technologies on their own.
Today’s distribution system architecture and mechanical control limitations greatly limit the potential
functionality provided by distributed generation. In addition to improved hardware, improved tools will be
needed for understanding and managing the interactions of distributed resources with existing distribution
systems, as well as developing control systems for large grids with a mixture of distributed and central
generation. As an example, to provide peaking power and premium power support for a distribution system,
distributed resources must be dispatchable. This will require adding a variety of remote monitoring,
communications, and control functions to the system as a whole. Moreover, distribution systems with mixed
distributed and central assets are likely to require dedicated volt-ampere reactive (VAR) generation for
85
TIAX LLC, U.S. Residential Information Technology Energy Consumption in 2005 and 2010, Prepared for U.S.
Department of Energy, March 2006, pp. 1-2.
86
Electric Power Research Institute, Electricity Technology Roadmap: 1999 Summary and Synthesis, pp. 33-34.
65
Chapter 6: Distribution Investment
system support and stability. In general, distributed resources will not produce VARs in the quantity or
location needed for grid stability. Distribution system operators will need the capability to produce VARs to
balance the system, either through the “must-run” generators of today or the “silicon VARs” of tomorrow.
The latter can be produced by the emerging family of High Power Electronic Controllers, which will use
power control devices to inject VARs into the system to stabilize voltage.87
Investments in Metering
Most electricity customers are served by conventional meters, which record cumulative energy usage and are
usually read once each month by a utility employee. Replacement of today’s electro-mechanical meters with
advanced “smart” meters will enhance customer service and customer options. Advanced, interval meters
measure power use on a time-differentiated basis and report via phone, Internet, or wireless. These meters
can track usage by the time of day, turn service on or off, diagnose problems, and react to price signals.
Digital power meters provide the ability to remotely monitor power usage and (increasingly) the ability to
perform other functions such as monitoring power quality, voltage, theft detection, remote
connect/disconnect, prepaid electricity purchases, and more. By collecting energy data on a real-time basis,
they will enable power companies to better understand consumption patterns and to work with customers to
cut energy usage.
As a result of this new technology, the meter will be transformed into a consumer gateway that allows price
signals, decisions, communications, and network intelligence to flow back and forth through the two-way
energy/information portal. This linchpin technology will help to create a more vibrant retail power
marketplace, with consumers responding to price signals and a variety of product options and choices not
previously available. The ultimate capabilities of an energy/information portal, in conjunction with an
automated distribution system, include: (1) advanced pricing and billing processes that would support realtime pricing; (2) consumer services, such as billing inquiries, service calls, outage and emergency services,
power quality, and diagnostics; (3) information for developing improved building and appliance standards;
(4) consumer load management through sophisticated on-site energy management systems; (5) load
forecasting; (6) long-term planning; and (7) green power marketing and sales.88
Installing new meters will be an expensive undertaking, however. Some experts estimate that about 10
million of the 130 million residential meters installed throughout the United States are equipped with
advanced technologies. Advanced meters cost approximately $100 to $150 per meter, so purchasing such
meters for 120 million residential customers would be an investment of approximately $12 billion to $18
billion.
87
88
Id., pp. 34-36.
Electric Power Research Institute, Electricity Sector, Framework for the Future, August 6, 2003, pp. 28-29.
66
Why Are Electricity Prices Increasing? An Industry-Wide Perspective
Minimizing Outage Costs
Power outages are very costly to retail customers and will become increasingly so in the future as more and
more applications require digital quality power. Even today, the nation’s industrial sector has become quite
dependent on high-technology processes. Air conditioning and other building climate control systems have
become more ubiquitous in the commercial sector, and the penetration of computers and other electronics has
increased throughout the economy. The value of electricity, or the cost of blackouts and other service
interruptions, correspondingly has increased.
Various approaches have been used to determine the value of reliability to customers. One method of doing
so is to assume that the value of having electricity is equal to the magnitude of the cost of not having it—i.e.,
the costs that a business incurs, in terms of lost sales, revenues, spoiled output, opportunity costs, etc.—as a
result of a power outage. A recent report prepared by ICF Consulting estimated the value of reliability by
computing costs of outages and other short-term reliability events.89 At an aggregate level, the study finds
that the annual historical value of outages and other reliability costs exceeds $20 billion per year and is much
higher in recent years. Moreover, this estimate excludes the costs associated with the August 2003 blackout
that affected much of the northeastern United States. This estimate also does not include the costs of very
short-term reliability events, such as voltage fluctuations, as ICF found little empirical data on this topic.90
However, the ICF study noted that some researchers found that momentary interruptions usually have a
higher per event cost than sustained outages. Hence, adding the costs associated with momentary outages
would significantly increase the $20+ billion estimated annual costs associated with sustained outages.
Indeed, one study estimated that approximately $52 billion per year is spent on momentary interruptions.91
Beyond reporting aggregate numbers, the ICF study also compared the value of electricity for residential,
commercial, and industrial customers to their actual prices paid. Previous studies estimated that the value of
electric service is approximately 100 times the price paid. The ICF report confirms this aggregate number,
but also sheds light on which sectors are more impacted in dollar terms by outages. For example, residential
customers value electricity the least in dollar terms, given their ability to react more flexibly to outages.
Industrial customers have the highest value relative to the low price they pay for electricity, but commercial
customers have the highest absolute value in dollar terms. This is because commercial customers are now
more exposed to more energy-intensive functions, while a large portion of industrial energy usage is
relatively less electricity dependent (i.e., is fueled by other sources).92
89
Bansari Saha (ICF Consulting), Value of A Reliable Supply of Electricity, Prepared for Edison Electric Institute, December
2005.
90
Id., p. 2.
91
Id., p. 18.
92
Id., p. 15.
67
CHAPTER 7
Environmental Investments
Overview
The electric power industry has been a focus of environmental regulation since the dawn of the modern
environmental protection era in the 1960s. Environmental protection has been, and will continue to be, a
driver of substantial investment in the power industry. Although such investments pay dividends in terms of
cleaner air, water, and land, they require substantial capital investment and increase operating costs—
expenditures that ultimately must be recovered in higher rates in order to maintain the financial integrity of
electric utilities. Under recently implemented EPA rulemakings, electric utility environmental costs are
expected to rise dramatically, with utilities planning about $40 billion in capital costs over the next decade
primarily to reduce air emissions. Enactment of additional, more stringent environmental rules could
substantially increase that level of expenditure.
The most important environmental issue for the electric utility industry is air emissions associated with
burning fossil fuels. The regulated pollutants—especially SO2, NOx, and recently mercury—are the focus of
substantial new requirements. Many utilities have participated in voluntary programs to reduce emissions of
CO2, the primary greenhouse gas contributing to climate change. However, mandatory programs to reduce
CO2 have been under consideration for some time, with regulatory activity emerging at the state level while
national policies are being actively debated.
Water resources are also critical to electric generation. Treatment facilities at power plants have reduced
direct releases of water pollutants to extremely low levels (in many cases below measurable levels). Thermal
electric plants (fossil fuel and nuclear power plants) use substantial volumes of water in cooling cycles, and
industry investments in cooling towers and other systems to dissipate heat before water is returned to its
source substantially reduce the impacts on aquatic resources. The industry faces significant new investments
to comply with recent rules requiring modification of water intake structures to minimize adverse impacts on
aquatic organisms. Moreover, there is growing concern in some regions of the country about the availability
of adequate water supplies, especially in arid regions experiencing significant population and electricity
demand growth.
Other environmental management costs arise in waste disposal (e.g., coal combustion products, including
scrubber materials), utilization (e.g., coal ash used in cement and concrete, scrubber by-products used in
gypsum board manufacturing), hazardous waste handling, and land management. Together, these
environmental expenditures are substantial and rising.
69
Chapter 7: Environmental Investments
Utility Environmental Protection Investments and Results
Largely as a result of investments in emissions reduction technology and policies that target existing
generating plants and other industrial sources, the air has become cleaner and will continue to improve
significantly even as overall electricity generation increases. Emissions of SO2 and NOx from electricity
generation have declined by nearly one-half since 1980, while electricity generation has increased by more
than 70 percent. As seen in Figure 7-1, most of this progress has occurred since the mid-1990s, as a result of
the acid rain provisions of the Clean Air Act Amendments of 1990 and subsequent programs to address
ozone transport in the eastern portion of the United States, as well as by the shift to cleaner generation
technologies such as natural gas combined-cycle plants. These reductions occurred mostly as existing plants
were retrofitted with new pollution controls and/or switched to lower sulfur fuels.
Figure 7-1
Historical and Projected Emissions and Net Generation
250
Historical
Projected
Index (1980 = 100)
200
150
NOx
NOX Emissions
so2
SO2 Emissions
Net Generation
100
50
2020
2018
2016
2014
2012
2010
2008
2006
2004
2002
2000
1998
1996
1994
1992
1990
1988
1986
1984
1982
1980
0
Source: EPA and EIA
Further reductions will also occur as a result of the implementation of the Clean Air Interstate Rule (CAIR),
which was issued by EPA in 2005 and requires additional reductions of SO2 and NOx emissions in the
eastern United States. In the West, the Clean Air Visibility Rule (CAVR) requires additional controls for
SO2 and NOx to reduce haze that affects National Park wilderness areas. The impact of these new rules on
emissions is also seen in Figure 7-1, based on EPA analysis.93 In fact, by 2020, electric generation is
projected to more than double from 1980 levels, while utility SO2 and NOx emissions will fall to one-quarter
of their 1980 levels. This means that overall electric industry emission rates (i.e., in pounds per MWh
generated) would fall by more than 85 percent—a remarkable technical achievement considering that most of
the coal-fired capacity responsible for 1980 emissions is projected to be operating 40 years later.
93
U.S. Environmental Protection Agency, Office of Air and Radiation, Multi-Pollutant Regulatory Analysis:
CAIR/CAMR/CAVR, October 2005.
70
Why Are Electricity Prices Increasing? An Industry-Wide Perspective
The Clean Air Mercury Rule (CAMR), which was promulgated in conjunction with CAIR, addresses
mercury emissions from electric generators for the first time. According to EPA estimates, mercury
emissions from electricity generation are projected to fall from about 50 tons per year currently to less than
30 tons per year by 2020, on a path to achieve an overall cap of 15 tons per year soon after. These
reductions occur both as a result of retrofitting existing plants for SO2, NOx, and particulate controls, as well
as installing specialized equipment such as activated carbon injection (ACI) directed at reducing mercury
emissions. Many states are considering more stringent and less flexible approaches to reduce mercury
emissions from power plants, which could reduce emissions faster but could cost considerably more than
EPA estimates for CAMR.
These projected emissions reductions are the result of a massive program of pollution control retrofits on
existing coal-fired capacity. In 2004, about one-third of coal units subject to the new rules had advanced
environmental controls; compliance with the rules described herein would raise that proportion to two-thirds
by 2020. On a capacity basis, the rules would increase the proportion of GW with advanced controls from
about 50 percent to almost 80 percent by 2020, meaning that about 200 GW of the expected 250 GW of coalfired capacity would have advanced environmental controls.94
These projected reductions in SO2, NOx, and mercury emissions are also consistent with a significant
expansion of new coal-fired generation, because new coal plants have very low emission rates compared to
older facilities as a result of technology advancements in emission controls and regulations that continually
reflect these advances. New coal-fired power plants must, at a minimum, meet the New Source Performance
Standard (NSPS) —technology-based emission rates that are revised only infrequently. In practice, new coal
plants must meet a Best Available Control Technology (BACT) or Lowest Achievable Emission Rate
(LAER) standard—standards that are tighter than the NSPS and determined on a case-by-case basis. The
BACT/LAER process explicitly considers technology improvements that make advanced pollution controls
more widely available and less expensive over time, and the NSPS is periodically tightened to reflect the
accumulated experience under the BACT requirements. The NSPS for coal-fired utility units was revised in
February 2006 and, as shown in Figure 7-2, mirrors the progress made in the BACT permitting program
since the early 1980s. Figure 7-2 also shows permitted emission rates for a proposed new IGCC plant, which
has extremely low emissions of SO2, NOx, particulate matter, mercury, and other pollutants.95
94
See U.S. Environmental Protection Agency, Office of Air and Radiation, Contributions of CAIR/CAMR/CAVR to NAAQS
Attainment: Focus on Control Technologies and Emission Reductions in the Electric Power Sector, April 18, 2006. The
250 GW capacity figure represents coal-fired units subject to CAIR.
95
Although several states have considered IGCC in BACT permitting evaluations of new pulverized coal-fired power plants,
thus far no state permitting agency has found that IGCC constitutes BACT primarily because of questions regarding
commercial availability and cost.
71
Chapter 7: Environmental Investments
Figure 7-2
Comparison of New Coal Plant Emissions Standards
Improving with regulatory evolution and technological innovation
Pounds per mmBtu
0.7
0.6
SO2
SO2
NOX
NOx
PM
0.5
0.4
0.3
0.2
0.1
0
1980-1984 BACT
permits
Recent BACT permits
2006 NSPS
IGCC
Sources and Notes:
Historical BACT permit data (1980 to 1984) found at http://cfpub.epa.gov/rblc/cfm/basicsearch.cfm, average of 28 permits for coal
units greater than 250 MW.
Recent BACT permit data (2000 to 2004) found at http://www.epa.gov/ttn/catc/dir1/natlcoal.xls.
2006 NSPS for SO 2 and NO X are stated at 1.4 and 1.0 lbs. / MWh, respectively, and are converted to lbs. / mmBtu assuming a heat
rate of 9.5 MWh per mmBtu (plant efficiency of 36%).
IGCC emissions rates are taken from Elm Road Generating Station permit dated January 2004.
Environmental progress has not been confined to air emissions, as utilities have reduced emissions of water
effluents to extremely low levels over time as their water discharge permits expire and are reissued with
more stringent requirements. This process has contributed to the continual overall improvement in U.S.
water quality experienced since the Clean Water Act was enacted in 1972. The most substantial new water
pollution compliance burden for utilities arises as a result of the Phase II program initiated under Section
316(b) of the Clean Water Act finalized in September 2004. This provision establishes technology-based
performance standards to minimize the adverse environmental impacts of cooling water intake structures at
existing plants. EPA estimates that 551 existing facilities will need to perform studies demonstrating
compliance with the standards, and that many will make costly operational changes and/or retrofits.
Beyond clean air and water, the nation’s electric utilities are involved in a variety of environmental activities,
ranging from waste disposal and recycling to pollution prevention and land management. These activities
are subject to a broad range of regulations at both the state and federal levels, and changes in these
regulations can increase utility costs. For example, the federal Spill Prevention Control and
Countermeasures (SPCC) program, which is focused on controlling oil and petroleum product spills from
fuel storage tanks, has been applied to oil-filled equipment at electrical substations, which potentially could
impose billions of dollars in new compliance costs on the industry.
Environmental Costs and Rate Impacts
All of this progress has entailed substantial costs to the industry, and costs continue to mount as compliance
deadlines loom. According to the most recent comprehensive national survey of environmental
expenditures, electric generators spent about $3.5 billion in 1999 for environmental compliance—almost 12
72
Why Are Electricity Prices Increasing? An Industry-Wide Perspective
percent of total industry environmental spending in the United States.96 The breakdown shown in Table 7-1
shows various expenditures by electric utilities and the corresponding figures for all industries. The
expenditures on air pollution control ($1.07 billion for capital and $0.91 billion in operating costs) together
comprised about 56 percent of total electric utility environmental spending in 1999. Although other cost
categories (not broken out in Table 7-1) are smaller, the electric utility share of these environmental costs
often is substantial. For example, U.S. utilities account for about 15 percent of non-hazardous waste disposal
costs in the United States, about 27 percent of site cleanup replacement, and 21 percent of overall industry
spending on habitat protection.
Table 7-1
TOTAL POLLUTION ABATEMENT AND CONTROL EXPENDITURES 1999
Section 1 - Pollution Abatement Capital Expenditures and Operating Costs
Industry
Segment
Air
Capital Expenditures (Million Dollars)
Water
Solid Waste
Multimedia
Total
Electric Power Generation
All Industries
1,071
3,464
55
1,802
17
362
2
182
1,145
5,810
% Electric
30.9%
3.1%
4.7%
0.9%
19.7%
Air
Water
Electric Power Generation
All Industries
910
5,069
100
4,587
127
2,013
27
196
1,164
11,864
% Electric
18.0%
2.2%
6.3%
13.7%
9.8%
Industry
Segment
Operating Costs (Million Dollars)
Solid Waste
Multimedia
Total
Section 2 - Other Types of Pollution Abatement & Control Expenditures
Industry
Segment
Electric Power Generation
All Industries
% Electric
Total Expenditures (Million Dollars), by Type
Disposal &
Pollution
Other
Recycling
Prevention
Expenditures
Capital Exp.
Operating Exp.
Total Exp.
Total Exp.
Payments to
Government
Total Exp.
46
399
406
4,924
443
2,768
213
3,155
107
959
11.6%
8.3%
16.0%
6.8%
11.1%
Section 3 - Aggregate Total of Pollution Abatement and Control Expenditures
Industry Segment
Total Expenditures (Million Dollars)
Electric Power Generation
All Industries
3,524
29,878
% Electric
11.8%
Sources and Notes:
Pollution Abatement Costs and Expenditures: 1999.
U.S. Census Bureau.
96
U.S. Census Bureau, Pollution Abatement Costs and Expenditures: 1999, November 2002.
73
Chapter 7: Environmental Investments
These 1999 figures do not reflect several substantial compliance burdens recently imposed on the electric
generation sector. These new outlays primarily involve additional investments at existing facilities to
comply with recent air emission requirements. According to an EEI survey of recent 10K reports, electric
utilities spent at least $3.2 billion in 2005 on environmentally related capital investments (compared to less
than $1.2 billion in environmental capital expenditures reflected in the 1999 figures cited earlier).
Environmental investments may be the fastest growing investment category in the industry over the next few
years.
EPA has analyzed the costs associated with CAIR, CAMR, and CAVR, and estimates that these regulations
will cost utilities about $3 billion per year in 2010, rising to more than $6 billion per year in 2020.97 The
overall net present value of all outlays between 2007 and 2025 is estimated to be about $50 billion, with
roughly one-half of that from capital investments in pollution controls. Moreover, the costs would be larger
under alternative legislative proposals that could be enacted, which feature more ambitious emission
reduction goals and, in some cases, less reliance on cap-and-trade emission approaches. Finally, many states
are adopting rules that go beyond federal requirements (especially in the case of mercury emissions), and
these programs will raise utility costs above the levels projected by EPA for compliance with the recently
finalized air rules.
According to EPA analysis, the Section 316(b) Phase II program will add an additional $400 million per year
in costs for electric generators, a substantial increase in the current level of expenditure on water quality.98
This figure could be higher depending on the degree of flexibility afforded utilities in actual implementation
of the rule, especially in the application of economic tests and the availability of restoration options.
Additional uncertainty regarding future water pollution control costs may arise as regulators implement total
maximum daily loads (TMDLs) that are specific to particular water bodies, which could necessitate
additional investments in wastewater treatment or other processes.
The costs of these new requirements are showing up in utility capital plans, as indicated in the EEI survey of
2005 10K reports. That survey revealed more than $40 billion in planned capital investments and other
environmental expenditures during the next 10 to 12 years, primarily to respond to recent air regulations.99 It
is important to note that these investments reflect compliance with current regulations as understood today;
additional requirements may arise as a result of legislation or new regulations. For example, health-based air
quality standards may be tightened for small particulates and ozone, and “reasonable progress” requirements
for improving visibility could trigger additional investments in pollution controls.
97
U.S. Environmental Protection Agency, Office of Air and Radiation, Multi-Pollutant Regulatory Analysis: CAIR/
CAMR/CAVR, October 2005. The precise figures are $2.7 billion in 2010 and $6.1 billion in 2020 (expressed in 1999
dollars).
98
U.S. Environmental Protection Agency, Office of Water, Economic and Benefits Analysis for the Final Section 316(b)
Phase II Existing Facilities Rule, February 2004, Chapter B-1 “Summary of Compliance Costs.” The exact figure cited in
this estimate is $385.5 million in annual pre-tax expenditures, including $196.2 million per year in capital costs (expressed
in 2002 dollars).
99
Because all utilities do not estimate their future environmental expenditures separately on their 10K reports, this figure
represents a conservative estimate of planned expenditures (although in cases where a range was given, the survey cited
the higher end of the range).
74
Why Are Electricity Prices Increasing? An Industry-Wide Perspective
An additional cost uncertainty arises from unresolved issues surrounding the implementation of the “New
Source Review” (NSR) and “Prevention of Significant Deterioration” (PSD) provisions of the Clean Air
Act—namely, the extent to which NSR/PSD applies when utilities undertake maintenance projects to restore
and/or improve availability, reliability, or efficiency of existing generating units. If NSR/PSD requirements
are deemed to apply to such maintenance projects, utilities could be required to install BACT emission
controls systems on that particular unit, regardless of whether the unit has recently installed controls, the unit
generally is uneconomic to control, or emission allowances generated from the operation of such controls
would simply be sold to other generators. Various enforcement cases and regulatory proposals are still
working through the legal system without a definitive resolution of this contentious issue, adding substantial
uncertainty to the costs of maintaining existing coal-fired generation capacity.
The financial impact of these outlays on utilities and ratepayers depends on the applicable cost-recovery
mechanisms, which vary by state and by company. Some states with traditional regulatory structures have
implemented specific mechanisms for full cost recovery of environmental compliance expenditures, while
others incorporate these costs into general rate cases, which can delay their recovery and create financial
stress. For states with deregulated retail markets, capital cost recovery is not assured, as generators depend
primarily on energy margins (i.e., market-clearing prices above their variable costs) for contributions to fixed
costs, which are largely unaffected by environmental controls. Here, compliance costs may simply go
unrecovered and result in additional financial stress on utilities during a period when they are making
investments to ensure adequate capacity and reliable service.
Climate Change and Electric Generation
The electric generation sector accounts for about 40 percent of U.S. CO2 emissions from energy
consumption, primarily as a result of the heavy reliance on coal-fired generation. In fact, coal-fired
generation, which is now 50 percent of total U.S. generation, accounts for about 33 percent of total U.S. CO2
emissions, or about the same portion as all transportation sources (motor vehicles, railroads, and aviation).
As a result, any mandatory policy to reduce CO2 emissions would fall heavily on the electric generating
sector. While energy efficiency, renewables, and new nuclear capacity will help to reduce the projected
growth in utility CO2 emissions, there currently are no economic technologies for CO2 removal from fossil
fuel-fired plants. Thus, in the near term, CO2 controls could entail a combination of fuel switching from coal
to natural gas (although this option has become much more expensive with increased natural gas prices),
additional renewables, stepped-up efficiency measures, advanced coal technologies, and perhaps additional
nuclear capacity. In the longer term, new technologies that could remove CO2 from fossil fuel generation
and permanently store CO2 underground are under development, but these are currently uncertain and
potentially expensive.
While it is extremely unlikely that mandatory CO2 controls will take full effect during the period examined
in this report, uncertainty over the eventual stringency, structure, and pace of potential CO2 emission
reductions adds significant risks for utility investment in new baseload generation. Given the size and scope
of the issue, coupled with the long expected lifetimes of generating facilities, the uncertainty regarding
possible greenhouse gas regulation may represent as much risk to utility supply planning as the uncertainty
regarding future fuel prices. As explained herein, some utilities already are taking steps to reduce CO2 and
75
Chapter 7: Environmental Investments
are incurring costs on a voluntary basis, in anticipation of eventual controls. If and when a CO2 control
policy is finalized, the industry will incur substantial “hard costs” of actual compliance.
Many utilities already are making investments that are influenced by the prospects of CO2 controls. The
electric utility sector has been a leader in voluntary projects to reduce greenhouse gas emissions or to
sequester CO2 emissions as recorded under the program authorized by Section 1605(b) of the Energy Policy
Act of 1992. In 2004, the electric power sector accounted for 1,489 projects, or 69 percent of the total
recorded under the 1605(b) program, which included direct reductions from generation, end-use efficiency,
cogeneration, and carbon sequestration. The 487 electric power and cogeneration projects provided an
estimated reported reduction of 173.7 million metric tons of CO2 equivalent (MMTCO2e) from direct sources
and 19.0 million MMTCO2e from indirect sources.100 According to an EEI analysis of the 1605(b) data, the
electric power sector accounted for a total of 282 MMTCO2e reductions, avoidances, and sequestration, or
63 percent of all reported emission reductions (assuming the higher figure of project-level reported
reductions and entity-wide reported reductions).101
In some states, integrated resource planning requirements stipulate that utilities consider CO2 policy risks.102
To the extent that resultant investment plans reflect the influence of potential CO2 controls, near-term costs
may increase. Likewise, some utilities have agreed to limit emissions in anticipation of eventual CO2
controls and associated costs. (Many of these utilities report their actions under the 1605(b) program
described above.) While such investments may well prove to be prudent and economic in the event that CO2
controls are instituted, they raise costs in the near term in order to manage the utility’s exposure to long-term
CO2 control costs.
The prospect for mandatory CO2 emission controls also is influencing capital markets and institutional
investors. Brokerage and investment rating firms, such as FitchRatings and Lehman Brothers, have begun to
incorporate the possibility of such regulations coming into force by the end of the decade into their outlooks
and planning, noting that such policies could have potentially significant impacts on the risks that the sector
faces. There has also been a noticeable increase in the filing of shareholder resolutions by some institutional
investors asking that electric utilities prepare reports discussing the potential impacts of such regulations on
future financial prospects.
100
101
102
Energy Information Administration, Voluntary Reporting of Greenhouse Gases 2004, March 2006.
Edison Electric Institute, Electric Power Sector § 1605(b) Summary for 2004, April 11, 2006.
Karl Bokenkamp, Hal Laflash, Virinder Singh, and Devra Bachrach Wang, "Hedging Carbon Risk: Protecting Customers
and Shareholders from the Financial Risk Associated with Carbon Dioxide Emissions," The Electricity Journal, July
2005.
76
Why Are Electricity Prices Increasing? An Industry-Wide Perspective
New Generating Technologies
Newer fossil fuel technologies are more efficient, and thus emit less CO2 per unit of generation than older
technologies. The conventional measure of generation efficiency is the “heat rate,” which expresses the
amount of fuel burned per unit of electricity generated (typically in Btu/kWh). Figure 7-3 shows CO2
emission rates for new generating technologies, based on their projected heat rates and fuels.
Figure 7-3
CO2 Emissions for Older and Current Generation Technologies
Emissions (tons CO2 / MWh)
1.2
Coal Technologies
Natural Gas Technologies
1
0.8
0.6
0.4
0.2
0
Older Coal Scrubbed Coal IGCC (without
Technologies
New
Sequestration)
[a]
Older CC
Technologies
[a]
Conventional
Gas/Oil CC
Advanced
Gas/Oil CC
Sources and Notes:
Assumptions to EIA Annual Energy Outlook 2006.
[a]: Assumed heat rates for older generation technology vintages = 10,000 btu / KWh for coal, 9,000 for CC .
Figure 7-3 includes emission rates for IGCC, an emerging “clean coal” technology that shows additional
promise in meeting CO2 reduction objectives because it presents a more economic opportunity to capture
CO2 emissions. The concentration of CO2 in the syngas produced in an IGCC plant is much higher than the
CO2 concentration of flue gases from conventional combustion systems, which could significantly reduce the
costs of capturing the CO2 for transport and sequestration in suitable geological formations, or deep ocean
disposal. This is an area of considerable research, and while estimates vary, it appears that carbon capture at
an IGCC plant would cost about half as much as capture applied to conventional pulverized coal plants.103
Among storage options being considered, expanding the use of CO2 for enhanced oil recovery may be cost
effective in some locations at current oil and technology prices, while other storage options could cost
anywhere between $1 per ton of CO2 to $20 per ton of CO2 per year on a levelized cost basis.104 These
capture and storage options are among the concepts pursued in the FutureGen Initiative, a collaboration
between the Department of Energy and private-sector interests to build and demonstrate the first CO2emission-free coal-fired plant in the world.
103
See Evaluation of Fossil Fuel Power Plants with CO2 Recovery, Final report prepared for U.S. Department of Energy,
February 2002. Also, U.S. Department of Energy, National Energy Technology Laboratory, Major Environmental
Aspects of Gasification-Based Power Generation Technologies, December 2002.
104
See Gemma Heddle, Howard Hertzog, and Michael Klett, “The Economics of CO2 Storage,” MIT Laboratory for Energy
and the Environment, August 2003.
77
Chapter 7: Environmental Investments
Nuclear power generation currently is the most significant non-CO2 emitting baseload generation
technology, and there is considerable interest in reviving nuclear construction, in large part because of the
role that nuclear could play in reducing CO2 emissions from the future generating fleet. Mandatory CO2
controls certainly will boost the economic prospects for nuclear power, and analysts expect that the costs of
new nuclear plants will fall as the first few units of the next generation of nuclear power plants are built,
which would help to further reduce the compliance cost burden of meeting CO2 goals. Although expanding
nuclear energy in the United States has economic risks and is constrained by the current impasse on highlevel waste disposition, it seems likely that nuclear will become an important element in long-term CO2
reduction policy.
Likewise, renewable electricity is largely CO2 emission free. (Emissions from biomass combustion are CO2neutral to the extent that they represent atmospheric carbon fixed in plant material through photosynthesis, a
process that can be repeated indefinitely.) The economic prospects of renewable energy would likewise be
boosted from mandatory CO2 emission limits, although their costs generally remain higher than conventional
generation without accounting for the CO2 benefit (and intermittent resources such as wind and solar pose
some operational challenges and incur additional costs as penetration levels increase in regional electricity
markets).
Costs of CO2 Controls
There are many estimates of the cost impact of potential CO2 controls on electricity suppliers, the results of
which vary based on the particular policy analyzed and the specific modeling assumptions employed.
However, they share one common feature: the impact on electricity prices would be roughly proportional to
the costs imposed on the industry. While climate change policies can be fashioned to reallocate cost burdens
in some respects—and thereby share some of the burden between utilities, ratepayers, and other sectors of
the economy—the costs imposed on electricity generation under most policies analyzed would dwarf the
amounts spent on clean air, water, and land described earlier. Electric power will still be produced and
consumed, and if electricity suppliers cannot recover their costs, the capital markets will extract additional
premiums on their debt and equity required to finance environmental controls and other infrastructure
investment. This either would inhibit capital formation when the industry would need to invest in new
generation technologies, or would show up in rates as additional returns needed to attract capital. This
means that mandatory CO2 controls would lead to price increases both through direct costs (such as
increased effective coal prices under a CO2 tax or cap-and-trade allowance system) and indirect costs, as
such policies would likely entail additional financial stress on utilities and raise the costs of capital.
78
CHAPTER 8
Financial Condition and Outlook
The previous sections of this report document the significant investment challenges faced by the utility
industry today. In the last few years, dramatic increases in fuel prices have driven the most significant
electricity price increases since the energy crises of the 1970s and 1980s. Because of factors such as longterm contracts, rate freezes, and deferred cost recovery, retail rates may not fully reflect these fuel cost
increases for another several years. The utility industry now also faces a wave of significant infrastructure
investment requirements that are driven by substantial capital needs for new generation to meet rising
demands, environmental compliance, expansion of the transmission grid, fuel diversity, and the continued
growth of the distribution system. In recent years, earned returns have been trending down as the increase in
utilities’ fuel and other costs exceeded growth in revenues.
This raises a most important question: does the utility industry have the financial strength sufficient to meet
the combined challenges of (1) sharply increasing and highly volatile fuel and purchased power costs; (2)
significant capital investment requirements; and (3) rising interest rates. The good news is that the industry
has recovered fairly well since the 2000 to 2002 financial meltdown that often is most vividly associated
with the western power crisis and the Enron bankruptcy. The bad news is that recent data also show a
downward trend in utility earned returns on equity (earned ROEs), a decline in operating cash flows, and
credit quality that has trended downward over the last five years. These findings suggest that reasonable rate
relief and investment recovery policies will be needed to maintain a financially strong utility industry
sufficiently capable of attracting the required capital and meeting its responsibilities in a stable, costeffective manner. Regulation that does not provide for the full and timely cost recovery of prudent costs will
weaken utilities financially, thereby raising investment-related costs and discouraging investments that
would yield long-term benefits.
The Industry’s Financial Condition During the Last Decade
Utility Credit Ratings
While primarily focused on assessing the risk of debt holders, credit ratings also reflect overall company and
industry fundamentals, as well as factors important to equity holders, such as allowed and earned ROEs.
79
Chapter 8: Financial Condition and Outlook
Figure 8-1 shows the credit ratings of electric and combination utilities, which are primarily utility operating
companies and reflect mostly the remaining regulated segment of the industry.105
Figure 8-1
Credit Ratings of Electric and Combination Utilities
100%
AA
80%
A+
AA
AA
AA
AA
A+
A+
A+
A+
A+
A
A-
A-
BBB+
BBB+
A
A-
BBB+
AA-
A-
ABBB
BBB+
BBB+
BBB+
BBB
BBB
BBB
BBB
BBB-
BBB-
BBB-
BBB-
BBB-
BB
BB
BB
1997
1998
BBB+
BBB
BBB
BBB
BBB-
BBB-
BBB-
BB
BB
BBBBBBBB
BB
Below
BB
BB
2002
20%
BBB
BBB
2001
BBB+
2000
BBB+
1999
30%
2005
A-
2004
A
A-
2003
A
A
40%
0%
A
BBB+
A
50%
10%
A
A-
70%
60%
A+
A
A
1996
Fraction of Rated Utilities
90%
Sources and Notes: S&P ratings as reported by Compustat. The sample consists of 121 companies based on Compustat GICS codes
for electric utilities and multi-utilities; to avoid double counting, it excludes holding companies if financial data for utility operating
subsidiaries is reported separately.
Figure 8-1 shows two notable trends. First, it documents the marked decline of average credit ratings for a
sample of 121 operating utilities, which represent the mostly regulated segment of the industry. Until yearend 1999, financially strong companies rated BBB+ or above accounted for approximately 75 percent of all
companies. By the end of 2005, the proportion of utilities rated BBB+ or above had declined to
approximately 45 percent. Second, Figure 8-1 documents that the financially weak segment of the industry
has been recovering from its weakest period in 2003. Utilities rated BBB- or below used to account for only
10 percent to 15 percent through 2001, but that share increased to almost 30 percent by 2003. Since then,
however, the proportion of utilities rated BBB- or below investment grade improved to approximately 20
percent by the end of 2005.
Figure 8-2 shows the same data for a sample of 25 independent power producers (IPPs) and energy traders,
i.e., the largely unregulated portion of the industry. Not surprisingly, this figure shows a much stronger
response to the recent energy and financial strain created in the aftermath of the western power crisis and the
Enron bankruptcy. The proportion of companies with below-investment-grade ratings (i.e., below BBB-)
increased from approximately 15 percent in 2000 to nearly 60 percent in 2002. As with utility operating
105
This sample contains 121 operating utilities. It consists of parent companies and subsidiaries for which financial data are
available and reported by Compustat, and includes only companies with Compustat GICS codes for “electric utilities” and
“multi utilities.” (This excludes companies in the deregulated segment of the industry, which are classified as
“independent power producers” or “energy trading companies.”) To avoid double counting and companies with sizeable
unregulated subsidiaries, we excluded utility holding companies whenever financial data for the utility operating
subsidiaries were reported separately.
80
Why Are Electricity Prices Increasing? An Industry-Wide Perspective
companies, the last few years show widespread improvement in overall credit ratings, though not to the
levels of the mostly regulated segment. By the end of 2005, the below-investment-grade-rated portion of the
industry still accounted for only approximately 40 percent of IPPs and energy trading companies. The
portion of IPPs and energy trading companies rated BBB+ or higher trended downward from more than 80
percent in 1996 to approximately 70 percent in 2000. The BBB+ and higher rated portion of this market
segment then declined to only 20 percent in 2002, before it recovered to a level of approximately 35 percent
by the end of 2005.
Figure 8-2
Credit Ratings of Independent Power Producers and Energy Traders
Traders
100%
A+
AA
A+
90%
80%
A
A-
A-
A-
A-
BBB+
BBB+
BBB+
A-
A-
BBB+
BBB+
BBB+
BBB
BBB
BBB
BBB
BBBBBB-
BB
BB
BBBBBB
BBB-
BBB-
BB
BB
BB
1999
BBB-
BBB
1998
BBB
BB
10%
BBB+
BBB
BBB+
30%
20%
ABBB+
BBB+
BBB-
50%
40%
A
A
A
A
A
60%
A+
A
1997
Fraction of Rated Utilities
A+
70%
A+
A+
Below
BB
2005
2004
2003
2002
2001
2000
1996
0%
Sources and Notes: S&P ratings as reported by Compustat. The sample consists of 25 companies based on Compustat GICS codes
for independent power producers and energy trading companies.
Earned and Allowed Returns on Equity
Financial data for the operating utility sample also show that utilities have been earning a median ROE that
exceeded the median of allowed ROEs in recent years. Figure 8-3 compares earned returns for the sample of
utility operating companies with allowed ROEs and trends in utility bond yields. The figure shows that in
the last several years, the median ROE for electric and combination utilities has been somewhat above
allowed ROEs. However, the figure also shows that earned returns already have been trending down as the
increase in utilities’ fuel and other costs exceeded growth in revenues. For 2003, 2004, and 2005, the
median earned ROE was only slightly above the median allowed ROE.
81
Chapter 8: Financial Condition and Outlook
Figure 8-3
Allowed and Earned Returns on Equity
For U.S. Electric and Combination Utilities
13.5%
13.5%
Earned ROE (Median)
Allowed ROE
(3rd Quartile)
12.5%
12.5%
11.5%
11.5%
10.5%
10.5%
Allowed ROE
(Median)
9.5%
Allowed ROE
(1st Quartile)
9.5%
8.5%
8.5%
7.5%
Earned ROE
(1st Quartile)
"Baa" Utility Bond Yield
7.5%
Sep-05
Apr-05
Nov-04
Jan-04
Jun-04
Aug-03
Mar-03
Oct-02
May-02
Dec-01
Jul-01
Feb-01
Sep-00
Apr-00
5.5%
Nov-99
5.5%
Jun-99
6.5%
Jan-99
6.5%
Sources and Notes:
Regulatory Research Associates, Compustat, Mergent Bond Record. Allowed ROE calculated as two-year rolling
average of commission-approved returns.
This downward trend in utility ROEs demonstrates that utility costs have started to outpace revenue growth,
suggesting further financial challenges ahead. But while utilities’ median earned ROEs are declining, they
are still (at least on average) within the range of allowed ROEs. So far, the decline in utility ROEs has been
mitigated partially by declining interest rates, as shown in Figure 8-3 by the trend in Baa-rated utility bonds.
Allowed ROEs also have declined with bond yields, although, as discussed below, utilities’ risks have
increased. As noted previously, this decline in allowed ROEs has raised concerns of rating agencies.
Figure 8-3 also shows that a sizable portion of the industry is earning returns that are well below investors’
required returns. One-fourth of utilities earn less than the earned ROE level shown with the line marked as
“1st Quartile.” This means utilities’ earned ROEs in this bottom quartile are significantly below the bottom
quartile of allowed ROEs and are, in fact, sometimes not much higher than the return on utility bonds.
Similar to what can be seen from the bottom range of utility credit ratings, this shows that, compared to the
“average,” a fairly sizable number of utilities are in a more vulnerable and relatively weak financial
condition. Importantly, since the earned ROEs of these utilities have declined more quickly than the ROEs
for the utility industry on average, regulatory policies that enable these utilities to recoup in a timely fashion
their rising fuel costs and to finance needed capital programs will be very important.
Increasing Risks
As discussed, the downward trend of utility credit ratings documents the increase in utilities’ average credit
risk, which raises their cost of debt. This means the decline in bond yields for Baa-rated utilities as shown in
Figure 8-3 is partly offset by the fact that utility credit ratings have been declining as well. A similar trend is
occurring with respect to utilities’ cost of equity. The risks to which equity holders are exposed have been
increasing, due to a variety of economic, operational, and regulatory factors. The increased risks mean that
82
Why Are Electricity Prices Increasing? An Industry-Wide Perspective
the risk premium required by utility equity investors has been increasing as well, which leads to higher
capital costs that also offset the general decline in interest rates.
The equity risk is commonly expressed through “beta,” which is a quantitative measure of the volatility of a
given stock price relative to the market as a whole. Figure 8-4 shows that the beta of the electric utility
industry has increased from approximately 0.55 in 2000 to approximately 0.85 in 2005. At a market risk
premium of 6.5 percent to 8.0 percent, this increase in risks raises the required ROE by approximately 2.0 to
2.4 percentage points (or 200 to 240 basis points).106 This increase in the required ROE approximately
offsets the decline in interest rates as reflected in the Baa-rated utility bond yields, as shown in Figure 8-3.
Consequently, the recent decline in allowed ROEs, as documented in Figure 8-3, may not be consistent with
the increase in utilities’ risks, as documented in Figure 8-4. Thus, credit rating agencies’ concerns over
“insufficient regulated authorized returns” also appear to be valid concerns from the perspective of equity
holders.
Figure 8-4
Trend of “Beta” for Sample of Electric Utilities (1995 to 2005)
1.0
0.9
0.8
0.7
Beta
0.6
0.5
0.4
0.3
0.2
0.1
2005
2004
2003
2002
2001
2000
1999
1998
1997
1996
1995
0.0
Source: Value Line Investment Survey. Average of reported betas for Value Line sample of 60 electric utilities.
Operating Cash Flows and Capital Spending
The magnitude of operating cash flows (or “funds from operations”107) relative to interest expense, total debt,
and capital spending frequently is used to assess the credit strength of companies. The size of operating cash
flows relative to a company’s interest expenses and other fixed obligations also indicates the flexibility that
106
This calculation is based on the “capital asset pricing model,” or “CAPM,” which says a company’s risk premium equals
the product of its beta and the market risk premium. In repeated empirical testing, the cost of capital turns out to be less
sensitive to changes in beta than the CAPM predicts. In particular, this research shows that low-beta stocks have higher
costs of capital and high-beta stocks have lower costs of capital than the CAPM predicts. Using this research, the
predicted change would be closer to 175 to 210 basis points, rather than 200 to 240 basis points.
107
The term “funds from operations” (FFO) is typically used by rating agencies. It is defined as operating cash flows
without adjusting the change in working capital. We are using FFO and operating cash flow interchangeably here.
83
Chapter 8: Financial Condition and Outlook
utilities have to withstand unexpected financial difficulties resulting from occurrences such as major plant
outages or storm damage. If operating costs increase faster than revenues, operating cash flows decline.
Importantly, trends in cash flows can be a leading indicator of utilities’ financial conditions because, unlike
earnings, cash flow cannot be preserved by accrual accounting and the deferred recovery of costs that often
occurs within the regulatory process.
Broadly speaking, the portion of capital expenditures that can be financed from internally generated funds is
equal to operating cash flows net of dividend payments. The extent to which funds from operations exceed
capital expenditures and dividends is defined as a utility’s “free cash flow” and measures the extent to which
utility companies need to rely on external financing.
As internal cash flow declines, a larger portion of a utility’s capital expenditures will need to be financed
externally, i.e., through the issuance of debt and/or equity in the capital markets. Unfortunately, it is not
always possible to “make up” declines in internal cash flows through external financing because access to
capital markets becomes more limited as a company’s financial flexibility declines. As documented by the
industry’s recent liquidity crunch, this ironically can lead to outcomes in which the companies that would
need to rely most heavily on external funds also find it most difficult to access such funds.
Figure 8-5 compares total operating cash flows (blue line) against the sum of capital expenditures and
dividends for the sample of utility operating companies. The figure shows that companies’ total operating
cash flows increased from approximately $35 billion in 2000 to approximately $45 billion in 2004. During
the same period, free cash flow improved significantly despite increased capital expenditures. These
improvements in cash flows again document the overall financial recovery of the industry in recent years.
Figure 8-5
Cash Flows of Electric and Combination Utilities
60
Capital Expenditure
Common Dividends
50
Net Cash Provided by Operating Activities
$ Billions
40
30
20
10
Sources
and
Notes:
The sample
sampleconsists
consistsofof121
121companies
companiesbased
based
Compustat
GICS
codes
electric
Sources
and
Notes:Compustat.
Compustat. The
onon
Compustat
GICS
codes
for for
electric
utilities and
utilities
and multi-utilities;
to avoid
doubleit counting,
it excludes
holdingif companies
if financial
for utility
operating
multi-utilities;
to avoid double
counting,
excludes holding
companies
financial data
for utilitydata
operating
subsidiaries
are
subsidiaries
is reported separately.
reported separately.
84
2005
2004
2003
2002
2001
2000
1999
1998
1997
1996
0
Why Are Electricity Prices Increasing? An Industry-Wide Perspective
Figure 8-5, however, also documents the more recent financial pressures that have emerged as utilities face
much higher operating costs and investment requirements. As the data show, from 2004 to 2005, operating
cash flows declined sharply, by approximately $10 billion, while capital expenditures increased. This
combination of reduced operating cash flows and increased capital expenditures foreshadows a likely further
decline in utility earned returns and significant financial challenges the industry is likely to face in the years
ahead.
A slightly different picture, but one that nevertheless suggests a similar outlook for the years ahead, emerges
for independent power producers and energy trading companies. Figure 8-6 shows that operating cash flows
declined from a high of approximately $17 billion in 2000 to approximately $15 billion in 2002 to 2004.
From 1999 to 2002, substantial capital expenditures associated with the construction of merchant generating
plants greatly exceeded operating cash flows. Similar to the electric and combination utilities represented in
Figure 8-6, however, operating cash flows have declined sharply in the last year: from $15 billion in 2004 to
approximately $10 billion in 2005.
Figure 8-6
Cash Flows of IPPs and Energy Trading Companies
35
Capital Expenditure
30
Common Dividends
Net Cash Provided by Operating Activities
$ Billions
25
20
15
10
5
2005
2004
2003
2002
2001
2000
1999
1998
1997
1996
0
Sources and Notes: Compustat. The sample consists of 25 companies based on Compustat GICS codes for
independent power producers and energy trading companies.
Summary: Utilities’ Financial Condition over the Past 10 Years
The past 10 years have been marked by several periods of retrenchment and improvement in overall financial
condition. The industry began the decade in 1996 in relatively good condition, with strong earnings and
credit ratings. Overall financial conditions did not change substantially until 2000.
To assess the financial ability of IOUs to meet their investment challenges, we first reviewed the evolution of
several key indicators of industry financial health during the past 10 years. This evolution helps lend some
perspective to a forward-looking financial evaluation later in this chapter. Our historic assessment examined
85
Chapter 8: Financial Condition and Outlook
four key indicators: industry credit ratings, earned and allowed ROEs, trends in utility risk, and operating
cash flows compared to capital expenditures.
As noted above, since 2003 the industry has been in a recovery mode. Although financial conditions have
improved substantially overall, the industry is still well below the financial measures of health experienced
in 1996 and longer-term historic norms. Furthermore, the lowest quartile of the regulated industry and a
larger segment of the unregulated industry continue to face conditions that require strong rate support.
From 2000 through 2003, the western power crisis and the following liquidity crisis in the energy trading and
merchant energy segment of the industry caused financial difficulties of nearly unprecedented scale and
severity throughout the industry. Only during and immediately after the energy crises of the 1970s and
1980s had the utility industry experienced a similar degree of financial upheaval.108 In the wake of this
financial crisis, unregulated industry participants sharply curtailed their capital spending, and even regulated
utilities struggled with severe financial pressures and an industry-wide credit crunch. In late 2002, the
president of the National Association of Regulatory Utility Commissioners (NARUC) and the chairman of
EEI pointed out in a joint statement:
“The electric power industry is now facing a financial crisis perhaps more acute than any
in its modern history. … All but a few electric power providers have found access to
capital increasingly costly and enormously difficult to acquire. … Left uncorrected, these
problems likely will further impede the financing and construction of critically needed
infrastructure, particularly high-voltage transmission and local distribution systems.
Significantly, this is a crisis affecting not just companies and their shareholders—
customers themselves and the U.S. economy are at risk if the industry cannot build out or
even maintain its generation and delivery infrastructure.”109
Financial Outlook: The Challenges Ahead
The previous discussion shows that the industry has been recovering reasonably well from the recent
financial crisis. The bottom end of credit ratings has improved somewhat for both utilities and unregulated
companies. Utilities’ earned ROEs are declining but, at least for the industry average, are still within the
range of allowed ROEs. Allowed ROEs have trended down, which has raised concerns of rating agencies,
but that decline in allowed ROEs is mitigated at least in part by declining interest rates and utility bond
yields. And, until recently, utilities on average had increasing operating cash flows that were sufficient to
fund most of the needed capital expenditure internally.
108
See, for example, Standard and Poor’s, “U.S. power company liquidity crunch: Déjà vu all over again,” December 16,
2002.
109
Joint statement of David A. Svanda, president-elect of the National Association of Regulatory Utility Commissioners, and
Erroll B. Davis Jr., chairman of the Edison Electric Institute, Chicago, November 13, 2002.
86
Why Are Electricity Prices Increasing? An Industry-Wide Perspective
These recent positive industry-wide developments do not imply that forward-looking conditions are expected
to remain as favorable, as several trends point to reduced industry financial strength. These trends include:
ƒ Utility earned returns have been declining as rate relief and revenue growth have been outpaced by the
combined effect of fuel and purchased power cost increases. In addition, the bottom quarter of the
industry is earning ROEs well below their cost of equity. In 2005, the earned returns of this segment
also have been declining at a more rapid rate. This suggests that a sizable portion of the industry may
be poorly positioned to address the challenges faced today and in the years ahead.
ƒ In 2005, operating cash flows declined more quickly than industry-earned ROEs, which likely is a
leading indicator of further earnings erosion.
ƒ The credit rating of the utility industry overall has trended downward to a point where, as of 2005, less
than half of electric and combination utilities were rated BBB+ or higher. In fact, over the last five
years, the typical utility credit rating declined from A to BBB. In addition, approximately 20 percent
of the industry is rated BBB- or below. The ability of these utilities to cope with additional financial
challenges may be very limited.
ƒ Further increases in fuel and purchased power costs, other increases in operating costs (including
labor, pension, and medical costs), the cost of complying with environmental and other regulatory
mandates, the additional capital costs of substantial infrastructure investment requirements, and the
possible future increase in financing costs (i.e., interest rates) will create a significant challenge to the
financial health of the industry in the years ahead. And again, while these challenges are significant
for the industry on average, variances across regions and companies will mean that a sizable number of
individual utilities will be affected much more strongly.
ƒ Finally, the recent sharp increases in costs have forced many utilities to file new rate cases, which
often are associated with delayed and sometimes incomplete cost recovery.
The following discussion addresses some of these issues in more detail.
The Outlook for Utility Credit Ratings and Earned Returns
Credit rating agencies have already taken note of the potential financial implications of the challenges facing
the industry today. According to one credit rating agency, Fitch, “unusually high and volatile natural gas and
energy prices raise risk overall.”110 Fitch further writes:
Volatile and rising energy commodity prices represent a challenge to investor-owned electric
utility companies. Many state regulatory commissions have approved procedures allowing
utilities in their jurisdiction to adjust tariffs periodically to reflect the actual cost of fuel and
purchased power. However, the plans in place for individual companies vary significantly in
their timing and effectiveness. Also, the implementation of rate adjustments is still subject to
regulatory and political risk, particularly in a period of rising energy costs. …A utility’s
ability to weather a period of high and rising commodity costs is influenced by many factors,
including the state’s market structure, rules regarding power procurement and the utility’s
110
FitchRatings, U.S. Power and Gas 2006 Outlook, December 15, 2005, p. 1.
87
Chapter 8: Financial Condition and Outlook
obligation to serve customers’ energy needs, the utility’s resource mix relative to its load
requirement, access to adequate liquidity and the state’s regulatory/political environment.111
In the 12- to 24-month outlook, potential negatives now loom larger than a year ago and are
no longer fully offset by the continuing benefits from the reduction in business risk that
resulted from the “back to basics” cyclical recovery, strong capital market access and low
interest rates. Taking a longer view, over the coming five years through 2010, the sector is
increasingly expected to face negative credit factors. These include rising interest rates,
higher capital expenditures and the continuation of volatile commodity prices.112
Other rating agencies and industry analysts have expressed similar concerns. While it clearly recognizes the
improvements in industry financial conditions over the last few years, Standard & Poor’s raises similar
concerns over the industry’s emerging challenges:
Certain measures of bondholder protection have stabilized following several years of gradual
improvement, reflecting debt reduction, divestiture of unregulated noncore assets, refinancing
of higher-cost debt, and tight cost control.
…While [2005 rating activity in the U.S. investor-owned utility industry was] more balanced
than in previous years, downside rating actions continued to overshadow upgrades. …
Downgrades in 2005 were attributable to overall deterioration in bondholder protection
measures, unsupportive rate decisions, heightened adversarial regulatory and political
development, burdensome construction programs, unrecovered investments, a focus on
shareholder value, and more aggressive growth strategies.
…Many companies face various business and financial pressures, which resulted in their
ratings going on CreditWatch with negative implications in 2005. CreditWatch listings and
rating outlooks are good indicators of prospective rating actions, and given the numerous
new and existing negative CreditWatch listings and negative outlooks, any upturn in overall
ratings quality is unlikely over the intermediate term.
…Much of the industry has been re-emphasizing its core competencies, but this is certainly
not without its own risk. These include the major pending regulatory decisions in certain
states, the need for substantial infrastructure expenditures, merger and acquisitions, fuel cost
recovery in a high-fuel-price environment, and still low, but gradually rising interest rates.113
111
112
113
FitchRatings, U.S. Electric Utilities: Credit Implications of Commodity Cost Recovery, February 13, 2006.
Id., p. 2.
Standard & Poor’s, Pace of U.S. Utility Rating Actions Picked up in 2005; Downgrades Dominate, February 1, 2006, pp.
1-4.
88
Why Are Electricity Prices Increasing? An Industry-Wide Perspective
These concerns over the challenges and financial health of the industry in the years ahead are not limited to
credit rating agencies and the perspective of debt holders; equity analysts similarly express concerns. For
example, Lehman Brothers states in a recent report:
Another Capex Cycle Looms – In this year’s study of electric utility regulation, we take the
results of a bottom-up compilation of fuel and capex spending increases over the next five
years and look at the implications for cash flow, returns, equity risk premia, and valuations.
We believe this analysis reinforces our negative stance on regulated electric utilities. …
Substantial Rate Increases – Infrastructure investments and high fuel costs spell rate shock,
demand destruction, and regulatory risk for traditional utilities. The projected 10 percentplus annual increases through the next four years could pain consumers, pressure politicians,
and harden regulators. …
Decreasing Returns – Historically, electric utility underearning coincides with free cash flow
turning negative (which happened in late 2005). …Our free cash estimates imply that earned
returns could drop to the nine percent ROE area in the coming years, a deficit of over 250 bps
versus projected allowed levels.114
Increasing Financing Costs
The capital costs associated with this clear need for infrastructure expansion will form another driver for rate
increases in coming years. While the current environment is quite favorable in terms of utilities’ access to
capital markets, recent increases in industry-specific risk factors and a trend to potentially higher interest
rates suggest higher financing costs for investment requirements going forward.
As shown in Figure 8-7, industry financing costs as measured by utility bond yields have reached a 40-year
low. As discussed earlier in this chapter, the decline in interest rates has allowed utilities to mitigate
increases in other costs. However, this long period of low and declining interest rates is not expected to
continue. During 2005, for example, the Federal Open Market Committee (FOMC) raised the Federal Funds
Rate a total of two percentage points. At its first meeting in 2006, the FOMC raised the Federal Funds Rate
another quarter percentage point to 4.5 percent. These recent increases in interest rates are also shown in
Figure 8-7, though they have not yet affected utility bond yields.
114
Lehman Brothers, Capital Lessons, Global Equity Research/North America, March 15, 2006, pp. 1-2.
89
Chapter 8: Financial Condition and Outlook
Figure 8-7
Utility Bond Yields vs. Federal Funds Rate (1/1960 to 1/2006)
20
18
16
Rate (%)
14
Bond Yield on Baa-rated Utility Issues
12
10
8
6
4
Federal Funds Rate
2
Jan-06
Jan-04
Jan-02
Jan-00
Jan-98
Jan-96
Jan-94
Jan-92
Jan-90
Jan-88
Jan-86
Jan-84
Jan-82
Jan-80
Jan-78
Jan-76
Jan-74
Jan-72
Jan-70
Jan-68
Jan-66
Jan-64
Jan-62
Jan-60
0
Source: Federal Reserve and Moody's.
The financing costs of utilities’ investment requirements are generally expected to increase. These increases
are likely for at least three reasons: (1) possible increases in long-term interest rates; (2) increases in utilities’
cost of debt due to declining credit quality; and (3) increases in utilities’ cost of equity due to higher risks.
In the earlier quotes, both Fitch and Standard & Poor’s specifically point to “rising interest rates” as one of
several negative credit factors faced by the industry going forward. However, while many industry analysts
anticipate that long-term interest rates will be increasing, the extent of such increases is still unclear. For
example, Lehman Brothers projects that the yield of 10-year government bonds in 2006 will have increased
by 90 basis points from 2005, without further increases through 2010.115 EIA projects an 80-basis point
increase from 2005 to 2006, with additional increases of 110 basis points through 2010.116 Based on the
long-range consensus forecast compiled by Blue Chip Financial Forecasts, bond yields are anticipated to
increase to 5.5 percent by 2009 and remain at that level for another five to 10 years.117 In comparison, as of
May 5, 2006, the yield on the 10-year government bond was 5.1 percent, which already exceeds May 2005
yields by approximately 100 basis points.118 These forecasts suggest that long-term interest rates in the years
ahead must be expected to be between 100 and 150 basis points above 2005 rates.
In addition to these trends of increasing interest rates, industry-specific risk factors will likely exert
additional upward pressures on utilities’ cost of capital. Rising operating costs, the evolution of industry
115
116
117
118
Id., p. 10.
Energy Information Administration, Annual Energy Outlook 2006, Table 19, February 2006 (based on projected yields of
AA-rated utility bonds).
Blue Chip Financial Forecasts, Long-Range Consensus U.S. Economic Projections, March 10, 2006, p. 15.
In May 2005, the 10-year constant maturity Treasury yield was 4.14 percent; during 2005, yields ranged from 4.0 percent
to 4.54 percent (http://www.federalreserve.gov/releases/h15/data/monthly/H15_TCMNOM_Y10.txt).
90
Why Are Electricity Prices Increasing? An Industry-Wide Perspective
structure, the ultimate costs of environmental and other regulatory mandates (see text box on following
page), and the extent and timeliness to which these costs can be recovered in rates introduce additional
uncertainties that are often difficult to quantify or hedge. These risk factors have already been recognized by
rating agencies through reduced credit ratings and negative outlooks. These risks will raise utilities’ cost of
debt relative to the general trend in long-term interest rates.
Similar upward pressures exist for utilities’ cost of equity. As the industry’s risks increase through factors
such as fuel price volatility, significant capital expenditures, regulatory lags, and the potential for incomplete
cost recovery, the required return on the equity-portion of utilities’ rate base will also tend to increase faster
than the general trend in interest rates. The increase in “beta” shown in Figure 8-4 indicates that utilities’
market risks today are already higher than in the recent past. Given the challenges ahead, these risks are
unlikely to decline.
In sum, fuel and market price volatility, along with uncertainties over the evolution of industry structure,
environmental costs, and timely recovery of these costs, as well as the required costs of financing significant
capital expenditures for infrastructure requirements, introduce operational and financial risks that are often
difficult to quantify or mitigate. This uncertainty itself is raising utilities’ financing costs at a time when the
sector’s capital needs are quite large.
91
Chapter 8: Financial Condition and Outlook
Financial Concerns Over Environmental Compliance Costs
And Imputed Debt of Long-Term Contracts
In addition to concern over the financial impact of increasing fuel, purchased power,
and infrastructure investment costs, exposure to large and growing environmental
compliance costs has become an important risk factor in the utility industry, both from
a debt- and equity-holder perspective. Shifting environmental requirements and
unclear technological solutions will only add to the financial concerns and uncertainty
over such environmental investments.1 The rating agencies have expressed particular
concerns over the current lack of clarity in environmental regulations, the added
financial burdens of significant investments required to comply with new rules, and the
risk that utilities may not be able to recover these costs in full or on a sufficiently
timely basis. The agencies also stated their fear that due to the combination of
growing compliance costs, high electricity prices, and an inconsistent and incoherent
regulatory approach to cost recovery, these cost recovery risks will not abate in the
near future—though mitigation opportunities exist in the form of environmental
financing and rate adjustment mechanisms.2 Similar opportunities exist to apply rate
adjustment mechanisms to mitigate adverse financial impacts of other regulatory
mandates, such as the financial pressures associated with the “imputed debt” of longterm purchases from renewable and other generating sources.3
1.
2.
3.
92
For example, see FitchRatings, Status of Environmental Regulation, October 12, 2004;
FitchRatings, Fitch Comments on EPA’s Clean Air Interstate Rule, March 16, 2005;
FitchRatings, Emission Trading, December 7, 2004; and Standard & Poor’s, Peer Comparison:
Three U.S. Power Giants’ Environmental Costs and Strategies, June 15, 2005.
For a discussion on mitigation of environmentally related financial risks, see Pfeifenberger and
Newell, “Innovative Regulatory Models to Address Environmental Compliance Costs in the
Utility Industry,” Newsletter of the American Bar Association, Section on Environment, Energy,
and Resources, October 2005, pp. 3-6. Also see, FitchRatings, New Missouri Bill Supports
Utility Credit, June 1, 2005, and Moody’s, Credit Opinion: Kentucky Utilities Co., June 2, 2005.
See direct testimony of Johannes Pfeifenberger re: state regulatory commissions’ rate treatment
of long-term purchased power costs. (Testimony before the Colorado Public Service
Commission, Docket No. 06S-234EG, April 14, 2006.) See also direct testimony of Michael
Vilbert re: implications and mitigation of imputed debt. (Testimony before the Public Utilities
Commission of Nevada, Docket No. 06-05__, May 31, 2006.)
CHAPTER 9
Cost Recovery, Investment, and Rates
In Perspective
In order for electric utilities to remain financially viable in the current era of increased operating costs and
continued need to invest in infrastructure development and expansion, rates must increase. Indeed,
electricity prices in many regions already have increased and further increases will be necessary in many
cases. For example, between January 2005 and January 2006, U.S. electricity prices increased by an average
of 11.6 percent, which predominantly reflected increased fuel and purchased power expenses. These
increases affected all customer classes: residential prices rose by 12.5 percent, commercial prices rose by
10.5 percent, and industrial prices rose by 12.6 percent. However, these increases were smaller than the
utilities’ increase in cost of fuel: the prices of fossil fuel consumed by utilities in 2005 were 30 percent
higher than those paid in 2004. 119
Historical Prices in Perspective
In many parts of the country electricity prices are increasing, sometimes by substantial amounts. Even with
these recent increases, however, electricity prices have risen less over the past few years than nearly every
other product or service Americans buy—and among energy products, much less. In many parts of the
country, electric rates have not gone up for many, many years. At the same time, most other products have
increased substantially in price in keeping with the rate of inflation.
Electric power continues to grow in value to American consumers and the American economy. As
electricity use has grown in economic value, its inflation-adjusted cost has been declining. From 1985 to
2000, average electricity prices rose 1.1 percent per year, less than half the average inflation rate of 2.4
percent. American homes use six percent more power today than they did in 1978. Yet even with 21 percent
greater use, the portion of our household budget that we devote to our power bill has declined.
Electricity Prices by Customer Class
Figure 9-1 illustrates the trends in average prices of retail electricity over the past 45 years by customer class.
What is striking about the figure is the relative stability in nominal prices until the first half of this decade.
Figure 9-2 provides the same results, but expressed in 2005 real dollars, and shows that in real terms prices
119
Energy Information Administration, Electric Power Monthly, April 2006.
93
Chapter 9: Cost Recovery, Investment, and Rates in Perspective
have been declining for the previous two decades. Thus, this observation helps explain how the cost of
electricity impacted consumers in general.
Figure 9-1
U.S. Electricity Prices by Class of Customer ($ Nominal)
10
Average Rate (¢/kWh)
U.S. Average
Residential Price
U.S. Average
Commercial Price
9
8
7
U.S. Average
Electricity Price
6
5
U.S. Average
Industrial Price
4
3
2
2005
2002
1999
1996
1993
1990
1987
1984
1981
1978
1975
1972
1969
1966
1963
1960
1
Sources: EIA Annual Energy Review 2004 and EIA Monthly Energy Review March 2006.
Figure 9-2
U.S. Electricity Prices by Class of Customer (Real 2005 Dollars)
14
Average Price (¢/kWh)
U.S. Average
Commercial Price
12
U.S. Average
Residential Price
10
8
U.S. Average
Industrial Price
U.S. Average
Total Price
6
2005
2002
1999
1996
1993
1990
1987
1984
1981
1978
1975
1972
1969
1966
1963
1960
4
Sources: EIA Annual Energy Review 2004, EIA Monthly Energy Review March 2006, and U.S. Bureau of Labor Statistics.
Note: Real dollars calculated from U.S. GDP deflator.
In order to place the cost of electricity to consumers in context, it is important to understand how such costs
have moved in relation to other key consumer products. Residential electricity prices in the United States
have increased at comparable or lower rates than other key indices of consumer prices. Table 9-1 provides a
comparison of retail electricity prices over time, as well as similar measures for other key consumer price
94
Why Are Electricity Prices Increasing? An Industry-Wide Perspective
indices. The top portion of the table shows the percentage change in electricity prices paid by customer class
over various time periods. The middle section analyzes price changes for other consumer energy products,
while the bottom part of the table presents core components of the Consumer Price Index (CPI). At any
point in time prior to the beginning of this decade, the pace of growth in electricity prices has been slower
than that of other energy products and core components of the CPI. For example, between 1995 and 2005,
rates for electricity increased by 17 percent, while the price of gasoline increased more than 95 percent and
inflation for all items increased 28 percent. The 2000 to 2005 picture, however, shows electricity prices
growing at a slightly greater rate than that of all items in the CPI. However, even in this period, other energy
prices are growing much more rapidly than electricity prices.
Table 9-1
Comparison of Electricity Rate Trends and Consumer Prices (1960 to 2005)
To see this more clearly, Figure 9-3 shows how nominal prices for electricity and several other major goods
have grown, using 1970 as a base year for an index. The figure shows that prices of medical care and natural
gas both have risen much faster than either electricity or gasoline, and consistently faster than the rate of
inflation. Like gasoline, electricity has at times risen faster or slower than the CPI. However, over the past
15 years, electricity prices rose much less than the general rate of inflation.
95
Chapter 9: Cost Recovery, Investment, and Rates in Perspective
Figure 9-3
Comparison of Electricity and Other Consumer Price Trends (1970 to 2005)
1,100
Natural
Gas
1,000
Base 1970=100
900
Medical
Care
800
700
600
Gasoline
500
400
300
Electricity
All Items
200
2004
2002
2000
1998
1996
1994
1992
1990
1988
1986
1984
1982
1980
1978
1976
1974
1972
1970
100
Sources: EIA Annual Energy Review 2004, EIA Monthly Energy Review March 2006, and U.S. Bureau of Labor Statistics.
How Electricity Prices Increase
Although changes in electricity prices in the United States have always reflected changes in underlying cost
drivers to some extent, the mechanisms by which operating costs and investments in infrastructure enter into
retail prices have become more complex and varied in the past decade. This is a result of regional and state
differences in rate regulation, wholesale market organization, generation mix, and the individual
characteristics of utilities themselves, such as their reliance on owned generation or purchased power to
serve load. The degree to which rising fuel costs translate into wholesale power cost increases also varies
quite substantially in regions with different market organization and generation mixes. In short, there is no
one price adjustment mechanism that applies universally to all regions, utilities, or their customers.
This situation has created widespread confusion for both regulators and consumers, as both the level and
timing of rate increases vary substantially from region to region. In some areas of the country, customers are
still protected by rate caps instituted during the 1990s and have seen little of the increased fuel or purchased
power costs reflected in their bills. In other areas, rate caps recently have expired, and sharply increased fuel
and purchased power costs have led to substantial price increases. In states where regulated retail utilities
operate with fuel adjustment clauses (FACs) that translate rising fuel and purchased power costs into rates,
customers have experienced increased rates and are bracing for more as the FACs are updated to reflect
recent fuel cost increases. Despite the variation in experience, it is clear that the fundamental cost driver of
increased fuel prices ultimately will increase electricity prices across the country, and that the character of
the price increases will have a substantial impact on the ability of utilities to pursue needed investment
priorities.
96
Why Are Electricity Prices Increasing? An Industry-Wide Perspective
The Role of Rate Increases
Without going into the particular mechanisms by which rates can and will rise, it is important to understand
the role of rates in the overall financial picture of retail electricity providers. Among other objectives, retail
rates must cover operating costs and maintain the financial integrity of utilities in order for utilities to pursue
needed investments, the benefits of which ultimately flow to ratepayers and society at large. Properly
designed and implemented, rates can also help utilities and customers to improve resource allocation.
Evolution in rate design to better improve the connection between wholesale market and retail prices can
reduce operating costs and rationalize investments in generating capacity and end-use efficiency or demand
flexibility. The fact that rates are rising should not divert attention from innovative rate structures that can
have benefits for both utilities and their customers.
The relationship between operating expenses, financial condition, the cost of capital, and the ability to pursue
investments is complex insofar as it represents a blend of current circumstances and future expectations
regarding how rapidly and accurately rates can change to enable utilities to fully recover costs, including an
adequate return on capital. One of the important aspects of adequate rates and viable rate adjustment
mechanisms is that they can and should lower investment costs of utilities. This means that investments
undertaken to enhance reliability, improve deliverability and power quality, expand generation capacity, or
improve environmental performance can be made at reasonable costs—and thus help to mitigate future rate
increases. In contrast, rates that compromise current operating cost recovery without adequate prospects of
adjustments will raise investment costs and discourage investment that can yield long-term benefits.
The Long-Term Benefits of Appropriate Rate Treatment
The previous chapters have detailed the potential benefits that arise from expanding investments in utility
infrastructure, and also outlined the current challenges that utilities face in raising external funds to increase
investment. This tension between near-term rate increases and the ability of utilities to invest in
infrastructure to help alleviate those cost increases in the long term is inherent in regulation. However, it is
worthwhile to point out that substantial benefits would occur if utilities are able to make the requisite
investments in infrastructure improvements. These benefits include:
ƒ Long-term reductions in operating costs, which would accompany investments in economic
baseload capacity, expanded transmission capability, and new distribution-level investments, which
could enable consumers to manage their energy costs more efficiently.
ƒ Enhancements of reliability and power quality, which would reduce the costs of power
interruptions and promote productivity based on the continuing penetration of digital equipment that
requires highly reliable service.
ƒ Improvements in competitive power markets, which would occur as transmission investments
enable more fluid and liquid wholesale power markets over broader geographic areas.
ƒ Cleaner generation from advanced coal, nuclear, and renewable power generation sources, which
also would reduce the costs of meeting potential future greenhouse gas mandates.
97
Chapter 9: Cost Recovery, Investment, and Rates in Perspective
ƒ Increased customer choice and control over energy use, which would arise from retail distribution
investments that enable innovative rate designs, distributed generation, and a host of technologies that
rely on enhanced connection between retail customers and wholesale power markets.
These are examples of the anticipated benefits from utility investments that could be imperiled by a backlash
against rising prices, which could lead to inadequate rate adjustments and an eroded ability of utilities to
pursue these improvements. While the near-term choices are often stark and unpleasant, the benefits from
prudent utility investment have not diminished in this era of high fuel costs and rising costs of capital. On
the contrary, the costs of inaction and capital investment atrophy are large and growing.
98
APPENDIX A
Household Power Use:
Past, Present, and Future
Figure A-1 shows how the typical household’s consumption of electricity has evolved over the last 25 years.
In 1978, more than 15 percent of a typical household’s electricity consumption was devoted to space heating.
Another 12 percent was devoted to water heating. By 2003, average household electricity consumption
increased 21 percent, from 1.07 kilowatt (kW) per hour to 1.30 kW per hour. The percentage of electricity
devoted to space heating and water heating declined, while the portion of electricity devoted to appliances
and air conditioning increased. These changes are the result of several factors, with the most notable being
the growth of personal digital appliances, such as computers, along with increased market penetration of air
conditioning and many other electric appliances.
Figure A-1 also shows projected household energy use in 2030. Average household consumption is
expected to increase by more than 11 percent, to 1.45 kW per hour. This increase will be entirely driven by
appliance-related consumption, largely reflecting the increased penetration of computers and other digital
technologies. The amount of electricity needed to heat water and living space, along with electricity needed
for refrigeration and clothes washing, is expected to decline slightly as these technologies become more
efficient over time.
Greater demand for electric power, however, does not translate directly into higher household expenditures.
To illustrate this point, Table A-1 and Figure A-2 show that the average American household’s total
spending on electricity has fallen steadily over time. Table A-1 illustrates that average annual expenditures
on electricity fell from 2.9 percent of total expenditures in 1984 to 2.5 percent of consumer expenditures in
2004.
Table A-1 also shows that this rate of spending has been lower than that of the other major categories of
consumer expenditures. Spending on housing, health care, and insurance premiums all more than doubled
over the 1984 to 2004 time period, while nominal spending on electricity increased by 69 percent. Since the
price of all goods rose by more than 82 percent during this period, inflation-adjusted household spending on
electric power declined.
99
Appendix A: Household Power Use: Past, Present, and Future
Figure A-1
Electricity Use in the Typical U.S. Home – Yesterday, Today, and Tomorrow
1978
Space Heating
0.17 KW / Hour
Air Conditioning
0.14 KW / Hour
All Appliances
0.63 KW / Hour
Water Heating
0.13 KW / Hour
Average Total Use: 1.07 KW / Hour / Household
2003
Space Heating
0.12 KW / Hour
Air Conditioning
0.19 KW / Hour
All Appliances
0.87 KW / Hour
Water Heating
0.11 KW / Hour
Average Total Use: 1.30 KW / Hour / Household
2030
Space Heating
0.11 KW / Hour
Air Conditioning
0.19 KW / Hour
All Appliances
1.06 KW / Hour
Water Heating
0.09 KW / Hour
Average Total Use: 1.45 KW / Hour / Household
100
Why Are Electricity Prices Increasing? An Industry-Wide Perspective
Table A-1: Comparison of Average Annual Consumer Expenditures
Figure A-2
At Today’s Electricity Prices, Electricity’s Share of the Household Budget
Is Smaller Than It Was 10 Years Ago.
Average Household Expenditures in 1994
Insurance and
Pensions
9.3%
Other
9.4%
Healthcare
5.5%
Electricity
2.7%
Housing (Excluding
Electricity)
29.1%
Insurance and
Pensions
11.1%
Other
9.2%
Electricity
2.5%
Housing (Excluding
Electricity)
29.6%
Healthcare
5.9%
Entertainment and
Apparel
10.1%
Food and Beverage
14.8%
Average Household Expenditures in 2004
Entertainment and
Apparel
9.3%
Transportation
19.0%
Food and Beverage
14.4%
Transportation
18.0%
Source: Bureau of Labor Statistics
101
APPENDIX B
Impacts of Price Increases on Electricity
Demand Growth Forecasts
Figure B-1 shows key inputs and outputs for EIA's most recent long-term forecast. Retail electricity prices
(in real cents/kWh) and costs for delivered fuel (in real $/MMBtu) are measured on the left axis. Projected
annual consumption (in billion kWh) is measured on the right axis. The forecast includes actual data for
2003 and 2004, while data for 2005 reflect the best estimates given the data available at the time of the
analysis.
Figure B-1 shows that significant increases in retail electricity prices in 2005 (that are now confirmed) were
included in EIA’s estimate. EIA shows an approximately 0.75 cent/kWh (+10 percent) real price increase in
2005, which is quite a significant one-year event. Also visible from the chart is a small flattening in the
slope of the demand curve between 2005 and 2006, indicating that some degree of price response of demand
is included in the forecast.
Thus, the historical calibration included in EIA's model shows sensitivity to recent spikes in retail prices.
Following this one-year price increase, however, EIA projects significant declines in the real price of
electricity, with a flattening in later years. This steady-to-declining trend in real electricity prices in 2006
and beyond closely tracks historical trends, and accordingly, demand follows a steady upward trajectory.120
In the context of EIA’s modeling framework, the fall in prices is likely due to several factors. First and most
important, both the rise and fall in electricity prices correspond closely with projected fuel costs. Second,
generating capacity additions underlying EIA's forecast are not dramatic in the near term, as EIA projects
about 50 GW of additions over the period through 2014, well below NERC’s forecast of 86 GW. Thus, the
rate base for generation is not growing at a significant pace in the near term under EIA's projections. While
we have no reason to doubt the internal consistency of EIA's projections and the underlying data, this
discussion suggests that such assumptions will impact the projection of demand growth. In EIA's
projections, fuel prices (notably natural gas) drop rapidly in price from a 2005 high, bringing electricity
prices down with them. With lower real energy prices, EIA is justified in projecting continued growth in the
120
EIA’s most recent Short Term Energy Outlook (May 2006) forecasts retail electricity prices to be essentially flat in real
terms from 2006 to 2008.
103
Appendix B: Impacts of Price Increases on Electricity Demand Growth Forecasts
demand for electric power. However, even with these relatively optimistic assumptions, EIA's demand
forecast is below NERC's latest base case forecast.121
Other Inputs
9
Retail Electricity Price
(¢ / kWh)
8
6000
Consumption
(billion kWh)
5400
4800
7
4200
6
3600
5
3000
Delivered Natural Gas Price
($ / MMBtu)
4
2400
2029
2027
2025
2023
2021
2019
2017
0
2015
0
2013
600
2011
1
2009
1200
2007
1800
2
2005
3
Consumption (billion kWh)
10
2003
Real Price 2004 $ (¢ / kWh or $ / MMBtu)
Figure B-1
EIA AEO 2006 Electricity Price Forecast and Other Inputs
Source:
EIA Annual Energy Outlook 2006.
To gain a better sense of the potential magnitude of the price response of demand, we have conducted a
simple sensitivity calculation of possible price effects on EIA's projections. EIA’s electricity forecasting
model includes a constant price elasticity for residential and commercial energy usage. For residential
electricity use, the elasticity used is -0.15. In other words, for a 10-percent increase in electricity prices,
residential demand for electricity should fall by 1.5 percent. The price elasticity reported for the commercial
sector is -0.25.122 Thus, EIA assumes that commercial users of electricity are more capable of reducing
demand in the face of swings in electricity prices. The demand responses to price changes are phased in over
a three-year period. Specifically, 50 percent of the demand impact occurs in the year of the price change, 35
percent of the impact occurs one year later, and 15 percent of the impact occurs two years later.
121
122
Other publicly available sources of demand forecasts do not provide as much detail about electricity price assumptions.
Global Insight, Inc.'s (GII) Summer 2005 U.S. Energy Outlook (published in August 2005) provides the most
comprehensive forecast of both electricity prices and demand outside of EIA. Some of GII's results are described in
EIA’s Annual Energy Outlook 2006 forecast comparisons and provide examples of the relationship between price and
demand forecasts (http://www.eia.doe.gov/oiaf/aeo/pdf/forecast.pdf). For example, GII's forecast of real electricity prices
by 2015 is 0.5 cents/kWh higher than that of EIA’s. GII's electricity sale predictions are 61 billion kWh lower than EIA's
for the same year. While these snapshots do not provide sufficient detail to determine if the full difference in demand
flows from price assumptions, they provide more anecdotal evidence that increases in retail price assumptions will impact
the path of demand forecasts.
Assumptions to the Annual Energy Outlook 2006, Residential and Commercial demand modules.
104
Why Are Electricity Prices Increasing? An Industry-Wide Perspective
EIA's forecasts of short-term price elasticities are in line with the very large published literature on electric
price impacts, but these elasticities are definitely short-term rather than long-term values. Electric demand
analysts long ago established that the effects of a real price increase are spread over approximately the seven
years following an increase. The total effect of the increase (i.e., the long-term elasticity) is generally
regarded to be on the order of -1.0, seven times as large as EIA's short-term value. Under EIA's framework,
long-term elasticities are incorporated implicitly through its bottom-up approach of modeling. That is,
consumers will shift their expenditures away from products that are associated with higher input prices in the
long run.
As discussed in this report, a variety of factors could lead to higher retail electricity rates. One example is
that natural gas prices might remain at higher levels than projected. As shown in Chapter 2, EIA's own Short
Term Energy Outlook now projects higher natural gas prices in the next two years than what was forecast in
EIA's Annual Energy Outlook 2006. A wide variety of factors could result in deviations from the latter
forecast, all of which could impact both price and consequently demand.
Our sensitivity analysis simply calculates the effect of a sustained real power price increase with the same
short-run EIA elasticity effects over a three-year lag. In particular, real prices are assumed to increase 10
percent between 2005 and 2006, and then no change in real price is forecasted through 2014. Demand is
then adjusted based upon the short-run elasticity factors from EIA and the difference in price from the
underlying forecast in a given year. For example, in 2005, residential real prices were 9.81 cents per kWh,
and EIA projects the price to fall approximately three percent by 2006 to 9.51 cents per kWh. In our
sensitivity calculation, the 2005 rate is increased by 10 percent to reach a level of 10.79 cents per kWh.
Thus, prices are 13 percent higher in 2006 than projected by EIA: 10 percent flowing from the assumed
increase in rates and three percent flowing from the removal of EIA's assumed drop in price. Since real
prices are 13 percent higher than modeled in 2006, the total demand reduction is about two percent given an
elasticity of -0.15 (.13 x -0.15 = -0.02). Fifty percent of the demand reduction is assumed to occur in 2006,
35 percent of it is assumed to occur in 2007, and the remaining 15 percent of it occurs in 2008, following
EIA's structure.
This calculation is preliminary and illustrative, but it highlights the linkage between prices and forecast
demand that can have a substantial impact on the amount and timing of new generating capacity needed. In
addition, demand is likely to be moderated through an expansion of demand-side management and demandresponse programs adopted by utilities.
105
APPENDIX C
Discussion of Historical Transmission
Investment Trends
There are several plausible reasons why transmission investment declined over the 1975 to 1999 period. One
reason is that utilities were building less generation during this period because many companies had ample
reserve margins and needed to “work off’ the excess generation constructed during the 1970s. These large
reserve margins tended to persist longer than expected because electricity growth slowed at this time. In
addition, to the extent new generation was being built, it tended to be smaller than the large baseload plants
that were built during the 1960s and 1970s. Moreover, the gas-fired units that became popular starting in the
mid-1980s could be added closer to loads and therefore had less need for incremental transmission than coalfired or nuclear units. One industry expert suggests that less transmission needed to be built because the
industry became more knowledgeable and skilled at getting more out of a given set of assets through better
system monitoring, pre-specified remedial actions to deal with operating contingencies, and automatic
protection measures.123 The industry’s financial condition also worsened starting in the mid-1970s, and it is
possible that construction budgets for transmission became a logical target for reductions. It also is possible
that industry management became more reluctant to invest in transmission because of uncertainty over the
rules governing cost recovery and use of the asset as the transition to a more competitive wholesale market
started. In summary, less transmission may have been built between 1975 and 1999 because: (1) it wasn’t
physically needed, and (2) utilities were less financially capable of making large investments in transmission.
While significant research would be needed to definitively prove or disprove these hypotheses, a review of
the data suggests that there is support for both of these primary reasons; i.e., less new transmission was
needed and utilities were less financially capable of making such investments. A review of the nationwide
reserve margin for IOUs shows that it jumped from 27.2 percent in 1974 to 34.3 percent in 1975, and stayed
above 30 percent through 1987. Indeed, in one year (1982) the reserve margin actually exceeded 40 percent.
Reserve margins for the period 1974 to 1996 are shown in Table C-1.
A review of the average plant size of new generating plants brought into service starting in 1980 confirms
the hypothesis that new plants started to get smaller at this time. Figure C-1 shows that the average size of
all new plants brought into service during the 1980s fell significantly throughout that decade and then
rebounded somewhat in the 1990s. It is interesting that the sharp upturn in average plant size that occurs in
the late 1990s coincides almost perfectly with the increase in transmission investment.
123
Phone conversation between Greg Basheda and Marty Baughman, May 4, 2006.
107
Appendix C: Discussion of Historical Transmission Investment Trends
Table C-1 Reserve Margins for IOUs (1974 to 1996)
Year
Reserve Margin%
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
Year
27.2
34.3
34.5
30.2
33.7
36.9
30.7
33.6
41.3
33.3
34.0
35.2
Reserve Margin%
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
32.8
30.6
25.0
28.6
25.6
25.3
26.7
20.7
20.1
15.2
17.6
Source: EEI, as reported in Leonard S., Andrew S. and Robert S. Hyman, America’s Electric Utilities: Past, Present and Future,
Eighth Edition, Public Utilities Reports, August 2005.
Figure C-1
Average Size of New Generating Plants
700
Average Size of Plants
that are Permitted or
under Construction
600
MW
500
400
300
Average Size of New Plants > 5 MW
200
100
2010
2008
2006
2004
2002
2000
1998
1996
1994
1992
1990
1988
1986
1984
1980
1982
Average Size of All New Plants
0
Source: Energy Velocity.
The financial aspect of this analysis is more complicated, because one would have to construct a “but for”
scenario of how much utilities would have invested in transmission under better financial conditions. It is
clear, however, that the 1975 to 1985 period was one of financial stress for the electric power industry. As
108
Why Are Electricity Prices Increasing? An Industry-Wide Perspective
Hyman et al. observe, the industry’s market-to-book ratio, as calculated by Moody’s, fell below 100 in 1974
and stayed below 100 until 1985 (when it reached 101).124 Thus, returns during this time were insufficient to
meet James Bonbright’s standard of profitability; i.e., returns should permit the utility to issue more stock at
prices not less than the per-share book value of the old stock.125 The 1975 to 1985 period clearly was not a
good time for the industry to raise significant amounts of capital.
These findings are important but not conclusive, because if the decline in transmission investment was due
primarily to ample generation and transmission capacity, changes in the size of new generators, and the
location of new generation investment, then the decline was likely to have little, if any, impact on
transmission reliability. However, if the decline also was driven by financial considerations, this suggests
that the decline could have adversely affected reliability as cash-strapped utilities decided to defer or scuttle
transmission investments. The first hypothesis seems to be the more likely of the two simply because there
is no evidence that the U.S. transmission system was inadequate, in terms of enabling reliable service, in the
last 25 years. None of the major blackouts that have occurred over the past 25 years has been found to be
caused by inadequate transmission capacity. Most have resulted from some combination of operator error
and inadequate maintenance practices (e.g., tree trimming). This is not to say that the transmission grid has
not become more “stressed” or “congested” over time; it has, as discussed in Chapter 5. But the system has
continued to meet long-standing industry reliability standards, which NERC and the Regional Reliability
Councils define in terms of minimizing the probability of incurring an involuntary outage or loss of load due
to a failure of the bulk power system.
124
125
Leonard S., Andrew S. and Robert S. Hyman, America’s Electric Utilities: Past, Present and Future, Eighth Edition,
Public Utilities Reports, August 2005, pp. 171, 189.
Id., pp. 170-171.
109
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement