52 |  ENERGY FUTURE: Think 

Efficiency American Physical Society • September 2008

Buildings
Commercial and residential buildings consume only one-
ninth as much petroleum as the transportation sector, but they
CHAPTER 3 still account for almost two-fifths of our nation’s primary
energy usage and more than one-third of our nation’s carbon
emissions. The technological potential for reducing energy
consumption and carbon emissions in the buildings sector
is considerable. However, as this chapter demonstrates,
significant progress likely will occur only if public policies
are adopted that address imperfections in the marketplace.
This is true even though the measures we suggest will result
in net savings for consumers.
This chapter identifies technologies and policies that can
increase the efficiency of energy use in commercial and
residential buildings. There are substantial opportunities for
achieving gains across the entire sector, including structures,
systems and appliances. In the case of commercial buildings,
a concept called integrated design provides extremely
fertile ground for research, development and demonstration
projects.

Introduction

A
mericans spend 90 percent of their time indoors, working, living, shopping and
entertaining in buildings that consume enormous amounts of energy.1 In 2006,
buildings — more than 118 million residential and commercial structures —
were responsible for 39 percent of the nation’s primary energy consumption, a level of
energy use that has a significant impact on global climate change and potentially on U.S.
energy security.2

Since most of their energy comes directly or indirectly from fossil fuels, buildings are
responsible for large quantities of greenhouse gas (GHG) emissions — about 36 percent
of the of CO2 associated with the nation’s total annual energy consumption.3 Building
energy consumption and the resulting GHG emissions, which have been steadily rising,
are projected by the Energy Information Administration (EIA) to increase another 30
percent by 2030.

Yet a large fraction of the energy delivered to buildings is wasted because of inefficient
building technologies. How much of this energy can ultimately be saved is an open
question — as much as 70 percent by the year 2030 in new buildings and perhaps more

1. In this report energy used by buildings includes energy used by building contents—appliances, vending machines, com-
puters, etc., the so-called “plug loads.” It also includes external loads, such as parking lot lights and swimming pools, that
use energy on building properties.

2. Sunlight, oil, natural gas, nuclear, coal, etc. are primary forms of energy–forms that are available on earth to be collected
and used. Electricity is a secondary form of energy—a convenient energy carrier that must be produced from primary en-
ergy.

3. All CO2 emissions are reported in metric tons (1000 kg). 1 Mt = 1 megaton = 1 million metric tons; 1Gt = 1 gigaton = 1
billion metric tons.
American Physical Society • September 2008 ENERGY FUTURE: Think Efficiency | 53

than 90 percent in the long term if there were pressing reasons to go that far. These energy savings can
be made not by reducing the standard of living, but by utilizing more efficient technologies to provide
the same, or higher, levels of comfort and convenience we have come to enjoy and appreciate. Some
of these technologies are available today; others are beyond our present grasp, but achievable in the
future with strong investment in research and development (R&D). Today significant energy can be
saved by making cost-effective efficiency improvements in buildings and their equipment—which
will reduce our nation’s energy consumption and GHG emissions and provide significant economic
savings to consumers.

Buildings consume 72 percent of the nation’s electricity, more than 50 percent of which is
generated from coal, our nation’s most abundant energy resource but one with CO2 emissions greater
than other fossil fuels, according to the EIA. The advantage of electricity is that it is a form of energy
that can be fully converted to work and is easy to distribute over the electric grid. Its disadvantage
is that it is generated and distributed with 31 percent efficiency—which means 69 percent of the
primary energy used to generate electricity is lost as waste heat before reaching the end user.

Building structures pose a more difficult problem than either the equipment they contain or
automobiles due to their long lifetimes and slow replacement rates. Whereas vehicles and appliances
wear out after a decade or so, buildings typically last for the better part of a century. Most buildings
were constructed during the years when energy was cheap, and as a result, they were not designed or
built with energy efficiency in mind. The overall number of buildings in the United States is growing
by only 1 to 2 percent per year. Hence a major reduction in building energy consumption must
involve both improvements in existing buildings and new construction.

Fortunately, widespread use of existing energy efficiency technologies and those that can be
developed over the near term would eliminate a sizable portion of the current waste of energy,
significantly reducing building energy consumption and greenhouse gas emissions. For the foreseeable
future, reducing primary energy consumption through improved efficiency is likely to remain far
cheaper than expanding renewable energy production [Glicksman, 2008]. (For a detailed definition
of energy efficiency. (See Endnote 1.)

Residential Buildings
In 2005, residential buildings in the United States consisted of 113 million residences totaling an
estimated 180 billion gross square feet, including standalone houses and mobile homes, as well as
dwellings located in apartment buildings and other multiresidence units [2007 Building Energy Data
Book, p. 2-1].
Figure 18
Residential energy end usage
In 2006 the residential sector consumed 21.8 quads4 of primary energy.
This chart shows the relative amounts going to various residential end uses.5
Electronics Cooking Computers
Space heating 6 Wet cleaning
Space cooling Water heating Lighting Refrigeration 1% Other

32% 13% 13% 12% 8% 8% 5% 5% 3%

20 40 60 80 100

Source: Energy Data Book (2007); EERE, U.S. Department of Energy

4. 1 Btu = British thermal unit, the amount of heat it takes to raise the temperature of 1 pound of water by 1 degree Fahrenheit. 1 quad =
1 quadrillion Btu = 1015 Btu. 1 Btu is also equal to 1054 joules, 1 joule being the metric unit of energy.

5. Numbers differ slightly from those in the DOE Building Energy Databook as the 4.7% adjustment has been eliminated and distributed
proportionally to all other categories.

6. Energy for “space heating” is the energy used to heat a building. Energy used to heat domestic hot water is included in the category
“wet cleaning
54 |  ENERGY FUTURE: Think Efficiency American Physical Society • September 2008

Data from recent Department of Housing and Urban Development surveys show that the average
rate of new construction is about 1.4 percent per year, and when demolition, condemnation, and
conversion of residences are factored in, the net growth per year is about 1.2 percent. Once built, a
residential building is likely to be usable for about one hundred years [Johnstone, 2001].

In 2000 (the latest year for which data are available), the average existing residential unit consisted
of 1,591 square feet and a household size of 2.7 people [2007 Building Energy Data Book, pp.2-1].
Although the housing market is currently extremely volatile, the trend for at least a half-century has
been toward larger residences. The average new single-family home constructed in 2006 was 2,470
square feet, 42 percent larger than in 1980 [2007 Building Energy Data Book, p. 2-3].

Primary energy consumption based on end use for residential buildings in 2005 is summarized in
Figure 18. The single largest end use is space heating (32%), followed by air conditioning or space
cooling (13%), water heating (13%) and lighting (12%). Note that these four combined account for
70% of the energy consumption.

Commercial Buildings
In the United States in 2000, 4.7 million commercial buildings provided 68.5 billion square feet.
From 2000 to 2005, the commercial building stock grew by 15 percent to 74.3 billion square feet,

0
Figure 19
Commercial energy end usage
In 2006 the commercial sector consumed 17.9 quads of primary energy. This chart shows the relative amounts going to various end uses.7
The category “Other” includes non-building commercial use such as street lighting, lighting in garages, etc.
Ventilation Computers
Water heating Cooking
8 8 Refrigeration
Lighting Space heating Space cooling Electronics Other

27% 15% 14% 7% 7% 6% 4% 3% 2% 15%

0 20

Source: Energy Data Book (2007); EERE, U.S. Department of Energy

double the growth rate of the residential sector [2007 Building Energy Data Book, p.2-5].

The commercial space breaks down as follows: offices (17%), mercantile (16%), education
(14%), warehouse and storage (14%) and lodging (7%), with numerous other functions making up
the remaining 32 percent.

Primary energy consumption based on end use for commercial buildings for 2005 is summarized
in Figure 19. The single largest end use is lighting (27%), followed by space heating (15%), space
cooling (14%) and water heating (7%).7 Together these four end-uses account for 63 percent of
primary energy consumption, somewhat lower than the case for the residential sector. Although
commercial buildings presently consume less primary energy than residential buildings, the energy
use in the commercial sector is experiencing nearly double the growth rate. (See Figure 20.)

7. Numbers differ slightly from those in the DOE Building Energy Databook as the 5.5% adjustment has been eliminated and distributed
proportionally to all other categories. Non-building commercial use includes electricity for street lights, water treatment plants, airport
lights, etc. All these numbers for energy uses should be viewed as rough approximations — more useful for comparing the relative sizes of
various energy uses than as precise figures of any specific energy use.

8. Since lighting and space cooling are predominantly accomplished with electric energy, their relative importance compared to other end
uses depends strongly on whether the focus is on primary (source) energy or site energy.
American Physical Society • September 2008 ENERGY FUTURE: Think Efficiency | 55

Primary Energy Figure 20

Figure 20 is a graph of
primary energy used by the
Total primary energy consumptions for buildings
Primary energy use (including that associated with electric use) for the
residential and commercial residential and commercial sectors in Quad (1015 Btu).
sectors from 1950 to the present Numbers after 2006
and projected out to 2030. The are projections.
graph indicates that energy 30
consumption in the commercial
sector is expected to grow faster 25
than that in the residential sector.
By 2030 combined primary 20

Quadrillion (BTU)
energy in the two sectors is Residential
expected to reach 51 quads, a
15
30 percent increase over 2006
consumption. 9
Commercial
Energy consumption has 30
been growing despite some
improvements in efficiency. 5
The main driving forces are
population growth and increased 0
standard of living associated 1950 1960 1970 1980 1990 2000 2010 2020 2030
with more and more ways to
Source: EIA 2008 Annual Energy Outlook
use energy. As compared with
30 years ago, Americans have
larger homes; more air-conditioners, televisions, and computers; and a variety of other devices that
use energy.

Currently available, cost-effective technologies could significantly reduce the energy consumption
of residential and commercial buildings, and the United States is making inadequate use of these
measures. But further technologically feasible advances could reduce consumption far more.

Using current and emerging technologies — those already in the pipeline – widespread construction
of cost-effective, zero-energy new single-family homes could be achieved in 10 to 15 years, except
possibly in hot, humid climates such as those in the Southeast. (By zero energy, we mean buildings
that use no fossil fuels. In general, that means reducing a building’s energy use by about 70 percent
from today’s average and fulfilling the remaining power needs with on-site or off-site renewable
energy.) Widespread construction of zero-energy commercial buildings will be harder to achieve,
but should be possible within 15 to 25 years, with a focused, sustained effort. Achieving 70 percent
reductions in energy consumption for new commercial buildings will require both new technologies
and greatly expanded use of the concept of integrated design. Such advances are unlikely to occur
without greatly expanded research, development and demonstration (RD&D) efforts.

R&D will also be needed to develop more ways to improve energy efficiency in existing buildings
through such measures as better wall insulation and windows.

But new technology alone will not assure efficiency improvements. Achieving maximum
efficiency in our nation’s buildings will require expanded use of policy tools such as appliance
efficiency standards, building energy codes and utility demand side management programs in order
to encourage efficiency.

8. Since lighting and space cooling are predominantly accomplished with electric energy, their relative importance compared to other end
uses depends strongly on whether the focus is on primary (source) energy or site energy.

9. Note that these projections do not include the impact of the Energy Independence and Security Act of 2007.
56 |  ENERGY FUTURE: Think Efficiency American Physical Society • September 2008

Clearly, reducing building energy consumption is critical to our nation’s future. A first step on the
path to limiting greenhouse gas emissions, reducing the national energy bill, avoiding unnecessary
construction of power plants and diminishing stresses on fossil energy resources is recognizing that
buildings (including factories) as well vehicles now consume vastly more energy than they need to
operate efficiently. (See Endnote 2.)

Finding 1:
If current and emerging cost-effective energy efficiency measures10 are employed in new buildings,
and in existing buildings as their heating, cooling, lighting and other equipment are replaced, the
growth in energy demand by the building sector could be reduced from the projected 30 percent
increase to zero between now and 2030. (See Endnote 3.)

Discussion:
There are a wide variety of technologies and strategies now available that can significantly
lower building energy consumption without any loss of service or comfort. Some are appropriate for
residential buildings, some for commercial buildings, and some for both. We are not suggesting that
all of these items are cost-effective in all cases.

Space heating is the largest residential user of energy, and cooling is a close second. Focusing
on those two systems, measures for both new construction and renovation that can save significant
amounts of energy include:

• Increasing insulation in walls, roof, floor and basement to cost-effective levels.

• Using window coatings, chosen based on climate, to reduce the amount of heat gain and
loss through thermal transmission.

• Moving heating and cooling ducts into the conditioned space (so that air from leaks is
not lost to the outside) for new construction and reducing leakage for new and existing
homes.

• Improving heating systems through the use of furnaces that send less than 10% of their
heat out the flue, variable-speed and higher efficiency motors/fans for air circulation and
efficient ground-source or gas-fired heat pumps.

• Upgrading equipment for cooling to achieve better heat transfer from an air conditioner’s
evaporator and condenser coils. Using variable-speed drives that allow units to operate
efficiently at partial load (rather than turning on and off frequently). In addition to saving
energy, this partial load operation also controls humidity more effectively and reduces the
internal heat loads on the air conditioner.

• Changing ventilation system installation (mostly for new construction) from the current
practice of relying on construction errors and accidental leakage to provide sufficient fresh
air to a process that uses the proper amount of mechanical ventilation while sealing the
home to nearly airtight standards.

• Controlling ventilation can mitigate problems with indoor air quality and mold, while also
recovering energy from the exhaust air stream.

10. By “emerging technologies,” we mean technologies that are likely to be available in the coming 5–10 years, assuming continuation of
the present level of R&D. By “cost effective,” we mean a technology that is cost-effective for the individual consumer. That is, at current
energy prices, the consumer would save more in reduced energy consumption over the lifetime of the technology than he would pay to
purchase and install the technology, assuming a real discount rate of 7 percent.
American Physical Society • September 2008 ENERGY FUTURE: Think Efficiency | 57

• Expanding use of evaporative cooling, using direct evaporation in arid climates, and
evaporation combined with an air-to-air heat exchanger in more humid climates.

• Constructing buildings with “cool” roofs that reflect rather than absorb infrared radiation in
warm and hot climates.

• Integrating passive solar heating and cooling into home designs. There are considerable
difficulties of custom-designing the orientation and thermal characteristics of individual
homes, but when it is done correctly, passive solar construction is a very cost-effective
measure for saving energy.

The remaining measures focus on the other high-energy end-uses: hot-water heating, lighting,
refrigeration, electronics/computers and other appliances.

In the residential sector, water heating uses as much energy as air cooling. This energy use in
all buildings can be cut by utilizing more efficient water heaters, reducing distribution losses in the
plumbing system, and reducing the heaviest demands for hot water in the home through water-saving
appliances (dishwashers and clothes washers).11

Experience indicates a great deal of energy can be saved through increasing the efficiency of
appliances. The best example may be refrigerators. Today’s refrigerators use one-fifth as much
energy as comparable refrigerators did 35 years ago. Also they cost less, after inflation. These energy
efficiency improvements have come about at least in part in response to federal regulations that require
manufacturers to meet appliance energy efficiency standards that are increasingly strict over time.

Additional equipment that will result in significant energy savings in commercial buildings from
available technology include:

• More efficient lamps, ballasts and luminaires.

• Improved glazing with lower heat loss and appropriate solar gain.
• Improved controls for air conditioning systems.
• Variable speed fans/drives and pumps.
• Lower-pressure fan systems.
• Occupancy sensors for controlling lights and ventilation.
• Efficient designs for building elevators and escalators.
Although analyses of energy savings stemming from single systems are the easiest to understand,
they miss many of the big-picture, cost-effective options that come from integrating systems, such
as:

• The use of lighting designs that optimize the distribution of light so that it is brightest
where the most light is needed and less intense elsewhere.

• The use of envelope designs that permit daylighting (described in the next section), while
controlling solar loads and glare.

• Reduction in size and/or complexity of heating, ventilation, and air-conditioning (HVAC)

systems made possible as a consequence of better insulation in walls, roofs, and floors;
improved windows; and reduced air leakage.

11. The reduced water usage arises not from changes in behavior, but from using dishwashers and clothes washers that provide the same
utility while using less hot water.
58 |  ENERGY FUTURE: Think Efficiency American Physical Society • September 2008

• The use of separate space conditioning and fresh air ventilation systems that allow oc-
cupants to control the systems based on need.

• Separate control of cooling and dehumidification, so that cooling systems can be sized
to address cooling alone.

Lighting and window energy-efficiency technologies and strategies are common to residential
and commercial buildings, though some lighting technologies are only appropriate for commercial
applications. Lighting presents perhaps the greatest opportunity for immediate, cost-effective energy
savings in buildings.

Incandescent lamps, a century-old technology, are the major source of light for residential
buildings despite converting only 5 percent of their electric energy into light. Electric energy is
generated and delivered to end-use sectors with an average of 31 percent efficiency. (See Endnote
2.) This means the overall efficiency of converting primary energy into incandescent light is only
1–2 percent. Clearly lighting is an area in which there is great room for improved efficiency. Figure
21 shows the status of lighting technology in the United States as of 2001.12

One widely available alternative is the compact fluorescent light (CFL), which uses one-quarter
of the energy of an incandescent bulb to deliver the same light intensity and quality. Mercury is an
environmental concern in all fluorescent lamps; however, replacing incandescent lamps with CFLs
releases less mercury into the environment than traditional light bulbs when the mercury released
through the burning of coal for electricity generation is taken into account (at current allowable rates
of mercury emissions) [U.S. EPA, 2008].

Figure 21
National lighting energy consumption
In 2001, the U.S. consumed 8.2 quads of primary energy (corresponding to 765 TWh13
of delivered electricity) for incandescent, fluorescent, high-intensity discharge (HID)
and light-emitting diode (LED) lighting technologies. Incandescent lamps remain
the dominant lighting technology in the residential sector.
Lighting energy consumption
by major sector and light source Breakdown of lighting energy
390 billion kWh used for lighting
Outdoor
stationary 8% in all commerical buildings in 2001
Industrial
14% Residential
27% Incandescent
Outdoor Fluorescent
stationary
Industrial High intensity
discharge

Commercial 51% Residential

LED (<.1%)
HID Commercial
22% 0 100 200 300 300 500
Incandescent
40% Annual energy consumption (TWh / year)

Sources: Navigant Consulting, Inc., U.S. Lighting Market Characterizations, Volume I,

Fluorescent National Lighting Inventory and Energy Consumption Estimate, Final Report for
38% Department of Energy, 2002

12. Presumably the expanded sale of CFLs since 2001 has reduced energy used by incandescent lamps, but we were not able to locate more
recent data.

13. Electric energy is typically measured in units of kilowatt-hours (kWh). 1 kWh = 3,600,000 joules. 1TWh = 1 terawatt-hour = 1 billion
kWh.
American Physical Society • September 2008 ENERGY FUTURE: Think Efficiency | 59

To get a sense of the rough potential of improving lighting efficiency, assume that all incandescent
lamps in use in 2001 were replaced by lamps that use one-fourth the energy, such as CFLs. The
annual electric savings would be about 240 TWh, corresponding to 2.6 quads of primary energy. (No
doubt some of these upgrades have been accomplished since 2001, particularly in the commercial
sector.) A more precise recent analysis of lighting upgrades found that annual electric energy could
be reduced by 120 TWh (1.3 quads primary energy) by upgrading residential incandescent lamps
and upgrading ballasts and lamps in commercial buildings [McMahon, 2007].

Lighting upgrades will accelerate due to the enactment of the federal Energy Independence and
Security Act of 2007, which phases in limits on the sale of incandescent bulbs. CFL sales are already
booming, with annual sales now at 400 million units compared to 50 million units just 5 years ago.
Solid-state lighting now being developed promises to produce lamps that double the energy savings
from CFLs.

Expanding the use of natural lighting—so-called “daylighting”—can save an estimated 30–60

percent in lighting energy in many commercial buildings [Loftness, 2004]. Daylighting uses sensors
and controls to adjust artificial lighting in response to changing natural light coming through windows
and skylights. Wal-Mart used this approach to upgrade lighting in its 2,100 stores worldwide with
energy savings that have a two-year payback in energy costs alone [Zimmerman, 2007].

Lighting energy can also be reduced by making better use of task lighting combined with sensors
and controls that deliver light at appropriate levels where and when needed. Ironically, commercial
buildings use about five times as much energy for lighting (per square foot) as do residential buildings,
even though residential buildings are used more at night. One of the primary reasons for this is that
residential buildings make better use of natural lighting and task lighting.

The rapid expansion of modern electronic equipment has resulted in homes and businesses
containing dozens of smaller electronic loads such as computers, printers, faxes, copiers, microwaves,
televisions, VCRs, DVD players and cable boxes. Many of these devices go into a standby mode and
continue to use power even when turned off. A recent study estimated that an average California home
contained more than forty products constantly drawing power. Together, those products consumed
nearly 1000 kWh/year while off or in a low-power mode [Meier, 2008]. This represents about 8%
of the average U.S. household electric energy consumption. Replacing such devices with Energy
Star ([Link] rated devices would significantly lower energy consumption,
particularly in standby or low-power mode.

––––––––––––––––––––––––––––––––––––––––––
We pause here to discuss combined heat and power (CHP) because it would enable buildings to
make more efficient use of electrical generation plants. However, unlike the technologies mentioned
above, CHP would require significant additional R&D to be practical in many cases. Also, CHP is
not assumed in reaching the 30 percent energy efficiency improvement cited in Finding 1.

In addition to energy efficiency in a building, an energy supply technology directly associated

with the building—combined heat and power (CHP)—represents a significant opportunity for
energy savings, yet one that remains largely unexploited in the United States. The electric power
sector discharges roughly two-thirds of its energy—nearly 26 quads annually—to the environment
in the form of low-grade heat. That low-grade heat is being lost at the same time residential and
commercial buildings are consuming 7.5 quads of natural gas to produce low-grade heat. Clearly a
great deal of energy could be saved if waste heat could be delivered to places that need it. It sounds
simple, but is very difficult to accomplish with centralized electric power stations. A few power
plants do capture this waste heat and distribute it in district heating systems, but those types of plants
are more common in Europe.
60 |  ENERGY FUTURE: Think Efficiency American Physical Society • September 2008

For U.S. buildings, existing CHP opportunities are mostly limited to large building complexes
such as those associated with colleges, universities and hospitals, which provide heating and cooling
from a centralized natural gas or coal plant. These plants have the opportunity to produce both
electricity and steam, with improved efficiency over plants that just produce heat or electricity. More
opportunities could present themselves if communities develop more compact land use patterns,
which is desirable from a transportation systems perspective as well. (See Chapter 2.)

CHP for individual buildings has been demonstrated using natural gas microturbines and fuel
cells,14 which generate both electricity and heat for space heating and domestic hot water. Balancing
the heat and electric demands proves challenging for a single building. For these technologies to
achieve widespread use, R&D efforts are needed to bring down the costs of microturbines and fuel
cells and to address a variety of technical and financial challenges [Marnay et al., 2007].

––––––––––––––––––––––––––––––––––––––––––
In determining what efficiency gains are possible with current and emerging technologies, it is
useful to start by looking at what is happening under current standard practices. Contractors focused
on energy upgrades to existing residential buildings achieve energy efficiency improvements ranging
from 15 to 35 percent by installing better and more efficient insulation, windows (in some instances)
and lights; by eliminating infiltration and duct leakage; by upgrading furnaces, boilers and air
conditioners; by replacing the power supplies that waste electricity when their devices are in standby
or low-power mode; and by replacing old appliances with newer, more efficient ones.15

Energy service companies (ESCOs) regularly work with larger commercial customers to perform
energy audits followed by upgrades in lighting, HVAC equipment and system controls, by which they
achieve cost-effective energy savings. We were unable to locate performance data for U.S. ESCOs.
In Berlin, Germany, however, ESCOs have improved the energy efficiency of 1,400 buildings by an
average of 24 percent at no cost to building owners and a profit to the ESCO that paid for the upgrade
[C40 Cities, 2008]. U.S. results are likely to be similar. Generally, it is easier to achieve efficiency
gains in new buildings than in existing ones.

Finding 1 is also based on an analysis conducted in 2000 as part of the Clean Energy Futures
study [Brown et al., 2001] and recently updated to determine the potential for improvements in
buildings [Brown, Borgeson and Koomey, 2008]. The analysis concludes that using currently available
technology upgrades as they become cost-effective for current and new buildings would result in a 30
percent decrease in the annual energy consumption by residential and commercial buildings in 2030.
(Endnote 3 explains why Finding 1 is worded more conservatively than the Clean Energy Futures
study.) It turns out that the reduction erases the projected increase in energy consumption for the
buildings sector, so that 2030 consumption by buildings could be the same as it is today.

Far more energy savings are technologically achievable, but not cost effective between now and
2030 for the individual consumer. Additional upgrades would be cost-effective if societal costs and
benefits were taken into account.
As discussed later in this chapter, even the cost-effective energy savings will not be achieved by
market forces alone; significant policy tools and incentives will be required. And the policy tools will
also likely result in unexpected improvements coming into the marketplace, as has happened in the
past.16
14. Buildings use several kinds of hydrogen fuel cells (phosphoric acid, molten carbonate, solid oxide and PEM) for generating electricity.
The hydrogen for these cells is extracted from natural gas with a reformer before going to the fuel cell.

15. Analysis provided by David Lee of the Environmental Protection Agency’s Energy Star program projects that Energy Star–recom-
mended cost-effective energy improvements to existing homes should yield efficiency increases ranging from 8 to 38 percent, with a 28
percent national average.

16. For example, refrigerators from 1975 to present; clothes washers from about 1990 to present (there were no standards and incentives
before that); and California’s reduction in home cooling energy [Goldstein and Hoffman, 2004].
American Physical Society • September 2008 ENERGY FUTURE: Think Efficiency | 61

Recommendation 1:
The federal government should set a goal that the U.S. building sector will use no more primary
energy in 2030 than it does in 2008. That goal should be reviewed every 5 years in light of the
available technology and revised to reflect even more aggressive goals if justified by technological
improvements. Achieving the goal will require that the federal government implement a set of policies
and programs such as those discussed later in this chapter.

––––––––––––––––––––––––––––––––––––––––––
Finding 2:
The goal of achieving significant levels of construction of cost-effective new zero-energy
commercial buildings by 2030 is not obtainable without significant advancement in building
technology and without the development and widespread adoption of integrated building design and
operation practices.

Discussion:
Zero-energy buildings (ZEBs), or “net-zero buildings,” are an attractive concept achievable by
merging efficient grid-connected buildings with renewable energy generation. The ideal is to use on-
site renewable energy sources, typically a photovoltaic (PV) array, to annually generate as much energy
as the building uses. A building, at times, buys energy from the grid while at other times, sells energy
back to the grid. A ZEB is one that annually sells as much energy as it buys, or more. ZEBs are being
built today, but are generally not yet cost-effective. Indeed, if cost and footprint are not constrained,
one can simply add whatever renewable energy sources are necessary to achieve net-zero energy,
no matter the efficiency of the building. But widespread construction of ZEBs requires that they be
cost-effective and that the renewable energy sources fit into the building footprint.17 Since efficiency
measures are much cheaper per unit energy than on-site renewable energy, both cost and footprint
constraint lead to the requirement that such buildings first be made very efficient. Efficiency is also
important to reduce energy consumption so that the required renewable energy sources can fit into
the building footprint. A 70 percent reduction in energy consumption (as compared with conventional
buildings) has been adopted as a consensus target for ZEB—though it is an estimate.

Various organizations, including the U.S. Congress (in the case of federal buildings), the American
Institute of Architects (AIA) and the State of California, have called for all new commercial buildings
to be ZEB by 2030.18 The AIA and California have established a 2020 goal for ZEB for all new
residential buildings.

Commercial buildings serve a large and widely varying set of occupants and needs. For example,
auditoriums and stores may at times be unoccupied, and at other times be crowded with hundreds of
people. Some buildings are no larger than small homes while others accommodate 60,000 football
fans or 20,000 office workers. And although there are examples of standardized commercial buildings,
the largest buildings are often “one-of-a-kind” buildings with specialized criteria. Comfort and health
require appropriate ventilation, heating, or more likely, cooling. Design engineers, rightly concerned
about liability, commonly design systems for the maximum occupancy, and these systems typically
waste enormous amounts of energy when occupancy is low.

There has been growing interest in the construction of green and energy-efficient commercial
buildings. The Leadership in Energy and Environmental Design (LEED) certification, introduced
in 2000 has rapidly grown in popularity and demand.19 Despite this growing interest there has been
relatively little progress in reducing energy consumption in new commercial buildings.

17. This may prove impossible for multistory commercial buildings, in which case off-site renewable energy sources may be required to
achieve net zero energy.

18. DOE’s Energy Efficiency and Renewable Energy (EERE) Building Technologies Program has set the goal of 2025 for ZEB commer-
cial buildings.
62 |  ENERGY FUTURE: Think Efficiency American Physical Society • September 2008

Information about hundreds of green commercial building projects may be found on the internet,
many with impressive claims about their projected energy consumption. But obtaining actual energy
consumption data for green commercial buildings is difficult. There are a growing number of LEED-
certified new commercial buildings (552 through 2006), and the public assumes they are energy
efficient, but the only study of their energy use is a recent New Buildings Institute review. The
Institute obtained energy performance data for only 21 or 22 percent, of the buildings [Turner and
Frenkel, 2008]. Of those, only six achieved site energy consumption levels per square foot that
were 70 percent below the average for all commercial buildings per square foot. Only three of the
buildings achieved that level of savings in primary energy consumption.20 Still, the New Buildings
Institute concluded that the LEED buildings it examined were 25 to 30 percent more efficient than
the average new commercial building, but not everyone would reach the same conclusion from the
data. Whatever their efficiency, these 121 LEED buildings consume more total energy per square
foot (either site or primary) than the average for the entire commercial building stock.

It should be noted that energy efficiency is but one of many criteria for LEED building
certification and credits for energy efficiency are awarded based on design simulations, not measured
building energy performance. There has been very little work on validating whether projections of
performance correspond to actual building performance; that is an area requiring further research.
What’s needed is a comprehensive system for rating building energy efficiency. More often than
not, constructed buildings actually use more energy than predicted by energy simulations performed
during the design process [Sacari et al., 2007]. This may be due to flaws in simulation tools; failures
in the design, construction or operation of the building; or energy intensive “plug-loads” that were
not included in energy simulations.21 Monthly energy bills cannot distinguish between energy used
by building systems (lighting, heating, ventilation, air-conditioning, etc.) and plug-loads. Monthly
energy bills for a very efficient hospital are likely to be higher than those for an inefficient elementary
school. Neither design energy simulations nor monthly energy bills provide the complete picture of
a building’s energy efficiency.
Very-low-energy commercial buildings are so rare largely because they are very difficult to
design, construct and operate. The biggest barrier is the complexity of the buildings and their HVAC
systems, and the important interactions between the various building systems and components.
Significant efficiency improvements have been achieved when all of these factors were taken into
account—using a process called “integrated design.”
Integrated design is a process in which all of the design variables are considered together, and
hundreds or even thousands of combinations are analyzed to arrive at the optimal design which meets
user requirements and minimizes energy consumption. The usual linear design process simply fails
to account for interactions between the various building components—and these can have important
energy and cost implications. (See Endnote 5.) For instance, the direct energy savings associated with
choosing a better window technology may not justify the cost – and the linear design process rejects
the upgrade. But the integrated design process goes on to determine that the window upgrade allows
a smaller, more efficient HVAC system—with total cost savings that justify the window technology
upgrade.
An experimental program run by Pacific Gas and Electric (PG&E) in the 1990s showed that
55–65% energy reduction could be accomplished using an integrated design approach [Brohard et
al., 1997]. But the process was time-consuming and hard to replicate. The six low-energy LEED
buildings offer further proof that 70% reduction in energy use can be accomplished. The challenge
is to develop easily-replicable design and construction processes that achieve such results cost-
effectively.
19. See [Link]

20. Study data were made available to us by Cathy Turner of the New Buildings Institute. Data for 98 buildings were sufficiently detailed
to calculate primary energy. Average site and source energy intensity for all non-vacant commercial buildings were obtained from the EIA
2003 Commercial Building Energy Consumption Survey (CBECS) database, and are 95 and 198 kBtu/sf, respectively.

21. Plug-loads are electric loads associated with equipment and appliances that are plugged into power receptacles, and not directly associ-
ated with the operating of the building itself. (Lighting and HVAC are not plug loads).
American Physical Society • September 2008 ENERGY FUTURE: Think Efficiency | 63

Although it is a crucial component of the solution, integrated design cannot guarantee low-
energy commercial building performance. Even the best-designed buildings, with well-thought-out
integrated systems, can suffer in their construction by contractors who lack the skills and experience
to implement the details faithfully. And facility managers may not know how to operate a new system
properly. A $100 home appliance comes with a setup and operating manual; many buildings do not.

Recommendation 2:
To achieve the ZEB goal for commercial buildings by 2030 the federal government should create
a research, development, and demonstration program with the goal of making integrated design and
operation of buildings standard practice. Such a program should be carried out co-operatively between
the federal government, state governments and electric utilities, with funding coming from all three
entities.

Since reducing energy consumption and carbon footprint is one of the most important goals for
green buildings, any green building rating system, such as LEED, should give energy efficiency
the highest priority, based in part on actual energy performance, and require reporting of energy
consumption data.

––––––––––––––––––––––––––––––––––––––––––
Finding 3:
The goal of achieving significant levels of construction of cost-effective zero-energy residential
buildings by 2020 is feasible, except perhaps for hot, humid climates. Most of the required technology
to compete with traditional housing is available but inadequately demonstrated. To achieve this goal in
hot, humid climates will require increased R&D to develop low-energy dehumidification and cooling
technologies and strategies.

Discussion:
Cost-effective zero energy homes are not available today, but there has been significant progress
in developing efficient single-family homes. Employment of cost-effective efficient technologies has
resulted in new, low-budget, single-family homes that use half as much primary energy as comparable
conventional homes [Norton et al., 2005; Christian, 2007]. And 80 to 90% reduction in energy used
for heating (though not total energy) has been achieved by passive solar homes22 in Germany, Austria,
Switzerland, Sweden and France [Schnieders, 2008].

The U.S. Department of Energy’s Building America program directly addresses the fundamental
problems of bringing energy efficiency to new residential buildings. The program provides technical
support for builders to construct very energy-efficient residential buildings at low or no increased first
cost to the consumer. Building America works with builders who are responsible for more than 50
percent of new residential construction in the United States. More than 50,000 competitively priced
houses have been constructed under the program, with an average energy use for heating and cooling
that is 30 to 40 percent less than that of typical new residences. DOE’s new Builders Challenge sets a
more ambitious goal of 30 percent savings in total building energy. Still, this program has a long way
to go to meet the ultimate goal of constructing and selling zero-energy houses by 2020.

Building America addresses two basic problems in commercializing zero-energy houses: assuring
the cost and energy performance of state-of-the art technologies and acquainting the building industry
with the techniques to build such houses. There is an R&D effort associated with this program
that supports the need to reduce costs, improve energy performance and address the cooling and
dehumidification requirements of hot, humid climates.

22. A passive solar home uses a well-insulated and tightly sealed thermal envelope along with very efficient windows to reduce heating
load, and meets much of the remaining heating requirement with solar heating.
64 |  ENERGY FUTURE: Think Efficiency American Physical Society • September 2008

The Building America approach is an effective way to create new markets for energy-efficient
housing. Funds to support more demonstration activities could speed up the process of commercializing
very-low-energy houses. Promoting Building America along with programs that show the value of
building energy codes and strict efficiency standards for appliances will produce very large gains in
energy savings in new houses.

––––––––––––––––––––––––––––––––––––––––––
Finding 4:
The federal government is not investing sufficient funds in R&D for next-generation building
technologies, for training building scientists or for supporting the associated national laboratory,
university and private sector research programs.

Discussion:
Federal funding is especially important in the building sector, which is highly fragmented and
consists largely of smaller firms that are unable to conduct R&D or have no economic incentive to
do so because of an inability to capture the benefits of R&D. Yet funding for energy efficiency R&D
for buildings, especially commercial buildings, has declined significantly.

In the 1980s, when levels of effort were much higher than today, federal R&D on energy
efficiency in buildings achieved notable success. A National Academy study [NAP, 2001] estimated
the economic benefits from advanced window coatings and electronic fluorescent ballasts to be $23
billion (in 2000 dollars). Both technologies resulted from federally funded energy efficiency R&D
efforts that expended far less than $23 billion.

Examples of research, development and demonstration that could enable the achievement of deep
savings for the majority of new commercial buildings include:

• Computer tools: Improved computer tools are needed to facilitate integrated design by ana-
lyzing interactions among building elements that affect energy use. In addition to continued
development of complex computer tools such as EnergyPlus, the simulation developed over
years by DOE, there is a need for tools that are simpler to use and appropriate during the
early stages of design when key decisions are made. These simpler tools need not be crude;
indeed, with the low cost of computing, complex programs like EnergyPlus could be made
much more user-friendly to meet this need. Such programs could also be used for building
labels.

• Monitoring and control technologies: Advanced technologies are needed to support diag-
nostics, fault detection and control in real time for a variety of building energy systems.

• More efficient building components: Among the needs are advances in air conditioning
and ventilation systems; advances in LED and conventional lighting and their controls;
advanced, affordable coatings for windows; envelope systems that optimize air transfer, wa-
ter transfer and heat transfer together on a climate-sensitive basis; and building-integrated
photovoltaic systems.

• Test facilities: Controlled experiments for commercial buildings in different climate re-
gions would benefit from the creation of test facilities. These facilities would allow tests
of advanced facades (walls, roofs and windows) coupled with innovative HVAC systems
and next-generation controls and monitoring. Such facilities are needed in different climate
zones: cold winter/hot summer; hot humid summer; and mild winter/summer.

• Demonstration programs: Demonstration programs showing that commercial buildings

can be built to use 70 percent less energy than current structures would encourage the build-
American Physical Society • September 2008 ENERGY FUTURE: Think Efficiency | 65

ing industry to pay more attention to integrated design and other energy efficiency practices.
Unlike demonstrations for residential buildings, such commercial demonstration programs
should be seen as R&D rather than straightforward commercialization of a process.

• Static insulation: Nanotechnology developed for direct energy conversion devices can also
be applied to create high-performance thermal insulation materials for various thermal sys-
tems. Materials with nanometer-sized channels hold the promise of reducing heat transfer,
which will open the possibility of a thin, rigid, high R-value (a measure of insulation effec-
tiveness) insulation panel for retrofit of interior surfaces of exterior walls. Such technology
could also be applied to improve the performance of foam and fiberglass insulation.

• Dynamic insulation: Nanotechnology has the potential to develop switchable insulations

in which the thermal conductivity could be varied by an order of magnitude. For example,
this type of insulation would allow interior thermal mass elements to be “charged” during
the evening by night cooling, insulated in the morning and then used during peak afternoon
periods.

• Lighting: Solid-state lights can be used to increase lighting efficiency and applied to tailor
lighting distribution to specific needs within a commercial building. They are potentially
twice as efficient as fluorescent lamps

• Windows: Current research is developing windows with high insulation values and selective
control of the solar spectrum. Advanced materials for coatings and frames have the potential
to produce window systems that achieve net energy gains during the winter and substantially
reduced air conditioning loads in the summer.

• Active building facades: Long-term R&D could lead to active building facades that can
modulate daylighting, solar gains and ventilation in response to monitoring of interior condi-
tions. For example, application of innovative materials and mirrored systems could distrib-
ute daylight much deeper into commercial building interiors and might lead to reductions in
lighting energy requirements by 50 percent or more.

• Advanced air conditioners and heat pump systems: Today’s systems operate at about
one-fourth of ideal efficiencies. R&D on systems optimization, heat transfer enhancement
and advanced controls can lead to much higher efficiency in space conditioning.

• Natural ventilation: Properly designed and operated natural ventilation systems can reduce
cooling loads in commercial buildings by 50 percent or more in many U.S. climates. Pre-
diction of air flow and thermal conditions in large, open-plan buildings is needed to assure
proper operation under a variety of climatic conditions.

• Energy performance data and analysis: Buildings will be increasingly monitored for their
energy performance. The creation of these data on a broad scale opens enormous research
opportunities to understand energy performance of buildings in the real world. Compilation
and analysis of these data is of great importance in informing policy and guiding R&D.

• Indoor environmental quality, health and productivity: Concerns exist that very-energy-
efficient buildings can degrade health and productivity of building occupants. R&D is need-
ed to identify when and if such problems arise from high efficiency and to establish measures
to mitigate adverse effects if they occur.
66 |  ENERGY FUTURE: Think Efficiency American Physical Society • September 2008

As a means to insuring that R&D on energy use in buildings is able to thrive over the long term, it
is essential to train this and future generations of building researchers and leaders among building
energy professionals in government and the private sector. For scientists and engineers, gradu-
ate programs with opportunities to pursue energy efficiency research need to be established and
expanded.

Recommendation 3:
The federal government should increase its investment in R&D to achieve the ZEB goal of 2030
for commercial buildings and 2020 for residential buildings. The current investment of somewhat
more than $100 million per year is considerably less in constant dollars than the research program
of 1980, which led to important innovations. The 1980 program in today’s dollars would be about
$250 million, and we recommend that funding for building R&D be increased to that level in the
next 3 to 5 years, after which it should be carefully reviewed. The review should determine the level
of continuing federal funding needed for the program to reach its goals, including examining what
technology is ready to go to market. One use of the additional spending should be to expand the
existing demonstration program for low-energy construction of residential buildings, along with
associated research, as noted in Finding 3.

––––––––––––––––––––––––––––––––––––––––––
Finding 5:
A wide range of market barriers and market failures discourage investment in energy-efficient
technologies.

Discussion:
If so many energy efficiency measures are cost-effective why are they not adopted? This question
has stimulated considerable discussion [IEA, 2007; NAP, 1992; Cavanagh, 2004; Goldstein, 2007].
Consider the barriers that inhibit adoption of cost-effective technologies—barriers faced by consumers,
manufacturers, builders, designers and suppliers of efficient products.

These include:

• Not knowing: The utility customer knows her total bill but not the contribution of the dif-
ferent appliances and the heating and cooling equipment, nor the thermal integrity of the
house. Policies such as Energy Star labels and appliance and building standards and labels
are essential to overcome this barrier. Even with labels, consumers may not always be aware
of highly efficient products on the market or be willing or able to calculate the payback from
an initial higher purchase price.

• Not caring: For most consumers, energy is a small cost compared with other expenditures.
For example, prior to 2002 typical TVs with remote controls used 5 to 7 watts of standby
power when turned off to permit the instant-on feature to function. In 2002 TVs were re-
quired to reduce standby power to 3 watts or less to qualify for Energy Star. On November
1, 2008, standby power must be reduced to 1 watt or less for new standalone TVs to qualify.
For the individual consumer, the reduction from 6 watts to 1 watt represents just a few dol-
lars in savings per TV per year. That sounds trivial, but applied to 300 million televisions
across the United States, it represents about $1 billion in electric savings. The cost of making
the improvement is small, so the manufacturer has a strong incentive to reduce the standby
power to 1 watt to qualify for the Energy Star label. But given the overall cost of operating
a TV, the consumer is not likely to care about the slight improvement in standby power ef-
ficiency.
• Split incentives: If the energy-using equipment or building is owned by a person who does
not pay the energy bill, there is little or no incentive to invest in efficiency. Landlords who do
not pay for energy, which is typical, are not likely to gain an advantage from installing energy
American Physical Society • September 2008 ENERGY FUTURE: Think Efficiency | 67

efficiency measures. In residential buildings, about one-third of all dwellings are occupied
by renters. Split incentives can also apply within a single company: Often the capital budget
for building improvements is under one manager while the operating budget is controlled by
another.

• Stalled demand for innovation: If manufacturers do not produce energy-efficient products,

consumers cannot purchase them. And if consumers do not demand energy efficiency, then
producers have little incentive to make their products more efficient. This “chicken and egg”
problem applies to appliance manufacturers as well as to builders and building designers.
The circle can be broken by policy decisions but is not likely to be resolved by market forces
alone.

• Reluctance to change: An important barrier to improved efficiency is inertia. For many

years, manufacturers produced appliances with little concern about energy efficiency. After
appliance standards were implemented, first by California in 1978, and then by the federal
government in 1990, electricity consumption by new refrigerators declined over a 30-year
period from 1,72523 to 498 kWh/yr while increasing considerably in size.24 The same phe-
nomenon occurred for other appliances, although to a lesser degree. Prior to the standards,
energy use had been increasing; for refrigerators it was increasing at 6 percent annually.

• Utility profits coupled to sales: Traditionally utilities (typically electric and natural gas
companies) have rate structures that connect their profits to energy sales—the more energy
they sell, the more money they make. This offers a disincentive for the utility company to
help customers become more efficient and use less energy. Yet utility companies are best
positioned to assist customers in identifying ways to improve energy efficiency. Establish-
ing rate structures in which utility profits are decoupled from sales removes one of the most
important barriers to energy efficiency.

To make the situation even more difficult, the design process itself provides disincentives to
incorporate energy efficiency into buildings. For commercial buildings, the lack of coordination between
engineers and architects, the payment of design fees that discourage integrated design (which adds to
design costs as it later saves in operational costs) and the lack of the required complex knowledge to
make the building energy efficient all discourage the use of the best—that is, integrated—approaches
to design and construction.

Not only do fragmentation and inefficient design processes provide justification for more federal
energy efficiency R&D, they also mean that innovative energy-saving products are unlikely to be
produced by manufacturers and thus will not be available to consumers. This problem in the building
industry accounts for the inability of the industry to develop first-rate tools for integrated design and
operation of buildings.

The example of fluorescent light ballasts makes clear the need for policies to promote energy
efficiency. Standard core-coil ballasts were far less efficient than newer ballasts. There was no difference
in performance between the two ballasts, and the payback period for the efficient ballast was approximately
two years at 1987 electricity prices. In short, the inefficient ballasts made no economic sense. Yet outside
of five states that had banned the standard ballasts, inefficient ballasts captured 90 percent of the market
in 1987. (The efficient ballasts cost an average of $4.40 more than the inefficient one—$15.40 versus
$11—and produced an average savings of $2.15 per year—hence the two-year payback.)

23. This includes manual defrost; the average for top freezer automatic defrost in that year was 2121 kWh/yr.

24. The refrigerator standard that produced the greatest savings—the 1993 standard—did not emerge in a vacuum but was informed by
successful Oak Ridge National Laboratory compressor research that demonstrated what was possible. This example illustrates how R&D
and policy tools work together to advance efficiency
68 |  ENERGY FUTURE: Think Efficiency American Physical Society • September 2008

It is worth noting that the largest portion of purchasers of fluorescent lights are managers of
commercial buildings, who might be expected to make purchases with high paybacks and be familiar
with technology as simple as fluorescent ballasts. But it took the passage of a federal ban through
a 1988 amendment to the National Appliance Energy Conservation Act of 1987 to move the market
away from the inefficient ballasts.

These barriers are not unique to the United States. They are observed all over the world. Even
developing economies and centrally planned economies are subject to the same failures.

__________________________________________________

Experience has shown that particularly in the case of buildings, even the best cost-effective
technologies are not readily adopted without policies to pull them into the market place. This may be
especially true for the buildings sector, where unnecessary energy costs that may make little difference
to the individual consumer can have large cumulative effects.

Below we discuss several policy tools that we believe should be part of a portfolio of efforts
to promote energy efficiency in buildings. The detail about how to apply these tools is beyond the
scope of this study, and this is not meant to be a comprehensive list. For example, we do not discuss
electric rate decoupling, which would enable utilities to make money from reducing consumption, as
mentioned above. Our main point is to emphasize yet again the absolute need for both research and
policy to make progress in energy efficiency.

Finding 6:
Among the most effective tools for increasing energy efficiency in buildings are building energy
codes, labeling, audit programs and tax and other incentives for the purchase of efficient technology.
For appliances, heating and cooling equipment and lighting, both mandatory efficiency standards
(e.g. for appliances), voluntary standards (e.g., industry consensus guidelines for lighting usage), and
energy labels (e.g., the Energy Star label developed and promoted by the Environmental Protection
Agency and DOE) have been effective. Utility demand-side management (DSM) programs that provide
incentives for energy efficiency have
been very successful.
Figure 22
Electric savings from California’s Discussion:
energy efficiency programs We limit our discussion to appli-
Annual electric energy savings in California since 1975 associated with ance standards, building energy
appliance standards, building energy standards and utility DSM programs. codes, and utility DSM programs,
15% of annual electricity
as those have been especially
use in California in 2003 effective in the United States. Figure
45,000
22 shows the impact of the three
40,000
programs in California, calculated
35,000
conservatively by the California
GWh per year

30,000 Energy Commission. Since the mid-

Utility efficiency
25,000
programs 1970s electric energy use per capita
20,000 nationally has risen steadily while for
15,000 Building California it has remained relatively
standards
10,000 flat. Today Californians use about
Appliance
5,000
standards
5,000 kWh per person per year less
0 than the average American. Appliance
standards, building energy codes, and
Source: Art Rosenfeld, California Energy Commission
American Physical Society • September 2008 ENERGY FUTURE: Think Efficiency | 69

utility DSM programs are estimated to be responsible for one-fourth [Sudarshan, 2008] to one-third
[Rosenfeld, 2008] of the difference.25

Appliance Standards
In 2000, appliance standards reduced U.S. electricity use by approximately 88 billion kWh,
2.5% of total U.S. electricity use. That same year, the standards reduced peak generating needs by
approximately 21 GW (roughly equivalent to 21 large power plants).26 Over the 1990–2000 period,
standards have reduced consumer energy bills by approximately $50 billion, with benefits being more
than three times the cost of meeting the standards [ACEEE, 2008].

By 2010, existing appliance standards are estimated to cut annual U.S. electricity use by 268 billion
kWh per year, and that figure is expected to grow to 483 billion kWh by 2020. This means expected
reductions of about 7 percent and 11 percent of projected electricity use in 2010 and 2020, respectively.
Peak electricity savings are estimated to increase from 72 GW in 2010 to 158 GW in 2020, and annual
carbon dioxide savings from 240 Mt in 2010 to 375 Mt in 2020. The net savings from these standards
approaches $300 billion [Nadel et al., 2007]. New standards adopted after 2008 could increase these
totals substantially. Preliminary estimates by the American Council for an Energy Efficient Economy
(ACEEE) are that new standards to be implemented by mid-2011 have the potential to increase annual
savings levels by another 190 billion kWh per year, increase peak demand savings by an additional 80
GW and cut annual carbon dioxide emissions by another 165 Mt by 2030. Figure 23 shows the effect
of the appliance standards on the efficiency of three major appliances.

Utility Demand-Side Management Directed at Customer Energy Efficiency

Demand-side management (DSM) programs are programs in which some central agency, often an
electric or natural gas utility, invests money to assist customers in becoming more energy efficient.
The investment may be in education programs or customer rebates to encourage purchase of more

Figure 23
Impact of standards on efficiency of 3 household appliances
Gas furnaces
Central air conditioners
Refrigerators
Effective dates of national standards
110
Effective dates of state standards
100

90
Index (1972 = 100)

80 Gas furnaces
75%
70

60 60%
Central air conditioners
50

40

30
Refrigerators 25%
20
1972 1976 1980 1984 1988 1992 1996 2000

Source: A. Rosenfeld, California Energy Commission; S. Nadel, ACEEE, in ECEEE 2003 Summer Study, [Link]

25. The bulk of the gap may be explained by California’s moderate climate and other structural factors, including shifts in industry [Su-
darshan, 2008]. Note that according to Figure 5 these programs account for about one-fifth of the gap.

26. The unit of power is 1 watt = 1 joule per second. Large nuclear power plants produce energy at a rate of roughly 1 GW = 1 gigawatt
= 109 watts
70 |  ENERGY FUTURE: Think Efficiency American Physical Society • September 2008

efficient appliances, or the agent may pay for the bulk of the efficiency upgrade, as in weatherization
programs for low-income customers. (See Endnote 4)

DSM programs involving customer energy efficiency have reduced growth in electricity sales
in the short run by providing financial incentives for energy efficiency purchases by consumers.
Utility DSM has also served to transform markets by aiding the commercialization of new energy-
efficient products.

Analysis of specific DSM programs has shown benefits greater than costs. For the nation, total
annual utility expenditures on customer energy efficiency from 1995 through 2006 have varied
from a low of $880 million in 1998 to a high of $1,700 million in 1995. DSM is returning as
a favored utility program, with expenditure in 2008 estimated to be higher than the 1995 level.
Levels are expected to continue to increase for the foreseeable future.

Building Energy Codes

Energy codes are adopted at the state or occasionally local level in the United States and are
enforced by local code officials at the city or county level. Most states follow national models
established by the two nonprofit organizations that write model codes, the International Code
Council and the American Society of Heating, Refrigerating and Air-Conditioning Engineers
(ASHRAE).

California is one of the states that has not followed these models; it has been a leader in building
energy standards that it develops itself. Energy codes adopted since 1975 in California reduced
peak power demand in 2003 by 5.75 GW while reducing electric energy use by 11 TWh/yr. The
economic value of energy savings is more than $30 billion or more than $2000 per household. The
electric energy needed to cool a new home in California has declined by two-thirds (about 2400
kWh/yr to 800) from 1970 to 2006, despite the fact that today’s new home is about 50 percent
bigger and is in a warmer climate as new development occurs farther from the coast. The California
energy code was revised in 2002, 2005, and 2008; each revision cut energy use by 10–15 percent
compared to the previous iteration. This is an annual rate of improvement of about 4 percent.

There is little federal involvement in establishing building energy codes. The federal government,
through the Department of Housing and Urban Development, sets standards for manufactured
housing and DOE provides modest technical assistance to the model codes organizations.

While energy codes are often thought of in a context limited to new construction, they also
save energy in existing buildings. When a new tenant moves into a space in a commercial building
and replaces the lights or the HVAC system, that action triggers the energy code requirements.
When a home is remodeled, the systems affected must meet energy code: Thus, a kitchen remodel
requiring changes to the electrical system in California triggers the need to meet the lighting efficacy
standards. A few localities also require retrofits at time of sale for both commercial and residential
properties.

Energy codes typically offer two methods of compliance: a prescriptive checklist approach
and a performance-based approach that relies on simulated energy performance of the proposed
building compared to a comparable reference building. The performance approach is preferred
overwhelmingly by builders in states where a usable method of calculating and displaying
performance is available, because it allows the builder to meet the energy goal at the lowest first
cost. Calculations software that is accessible for use by architectural and engineering firms and
consulting companies that provide technical expertise in meeting codes is available nationally
for homes, but only widely in California for commercial buildings. The European Union (EU)
has requirements that a building energy label be developed for all new buildings and that energy
evaluation needs to take place when a building is sold. These will result in easily used software
throughout the EU. Efforts are under way to harmonize this development within EU member states
and with the residential system used in the United States.
American Physical Society • September 2008 ENERGY FUTURE: Think Efficiency | 71

Recommendation 4:
DOE should promulgate appliance efficiency standards at levels that are cost-effective and
technically achievable, as required by the federal legislation enabling the standards. DOE should
promulgate standards for all products for which it has been granted authority to do so, including
those appliances for which there is not a specific congressional mandate. A streamlined procedure is
needed to avoid delays in releasing the standards.

Recommendation 5:
Considering the cost effectiveness of utility DSM to date, and the fact that many states have
hesitated in creating such programs, the federal government should encourage states to initiate DSM
programs through their utilities. The federal role could be to provide rewards to states that have
significant and effective DSM programs and disincentives to those that do not.

Recommendation 6:
Building energy standards, such as those promulgated in California, should be implemented
nationwide. States should be strongly encouraged to set standards for residential buildings and
require localities to enforce them. For commercial buildings, performance-based standards that rely
on computer software to compare a building design with a reference building are implemented only
in California. The federal government should develop a computer software tool much like that used
in California to enable states to adopt performance standards for commercial buildings. States should
set standards that are tight enough to spur innovation in their building industries.
72 |  ENERGY FUTURE: Think Efficiency American Physical Society • September 2008

Chapter 3 Endnotes
Endnote 1. Energy Efficiency
Energy efficiency is traditionally defined as the ratio of the “useful energy” to the energy consumed
or taken in.27 A typical coal electric power plant takes in 100 units of coal energy and produces 34 units
of electric energy, making the plant 34 percent efficient. As noted in the introduction, an 80 percent
efficient natural gas furnace delivers 80 units of heat (useful energy) to a house for every 100 units of
natural gas energy consumed, with the remaining 20 units of energy lost as exhaust through the flue.

This traditional definition of energy efficiency is not adequate in identifying many opportunities
for reducing (primary) energy consumption through improved technology and alternate strategies.
Consider, for instance, two otherwise identical houses, one having no thermal insulation and the other
being well-insulated. Both are heated by 80 percent efficient natural gas furnaces. Let us suppose the
insulated house uses one-fifth as much energy for heating as does the uninsulated house. We view the
insulated house as being more energy efficient—but in what sense? In both cases the energy used to
heat is considered “useful energy”; hence both homes by our traditional definition are 80% efficient.
But the uninsulated house uses 5 times as much “useful energy.”

Another problem with the traditional definition arises when we consider electric resistive heat.
Electric resistive heaters are, in the traditional sense, 100% efficient at converting electric energy
into heat (useful energy). Hence replacing the aforementioned natural gas furnace with electric heat
would appear to be an improvement in energy efficiency. Yet if we trace the electric energy back to its
primary sources we see that the electric-heated home uses considerable more primary energy—and is
not to be regarded as more efficient.

Here we adopt a more general definition of energy efficiency that avoids these problems. This
definition of energy efficiency is the ratio of the minimum primary energy required to perform a task
divided by the actual primary energy consumed by the specific process.28 This second definition tells you
how well you are doing as compared with the best possible solution. For the electric power generation
from coal described earlier this yields the same 34% efficiency. We note that it is frequently difficult
or even impossible to determine the minimum primary energy required to accomplish a certain task.
But even without knowing that number, we can compare the efficiencies of two different methods
of performing the same task and determine their ratio. Applying this more general definition to our
earlier example of two houses, we find the insulated house to be 5 times as efficient (with respect to
heating) as the uninsulated house.29

With this new definition of energy efficiency it is instructive to consider again the typical natural
gas furnace, providing heat to keep the inside of a house at 70 ºF when the outside temperature is 32º F.
The task that this furnace achieves is use of natural gas to deliver heat to the interior of a house. What
is the minimum primary energy required? You could start with natural gas to produce electricity with
greater than 50% efficiency, then use the electricity to run a heat pump that pumps heat into the house
from the outside air or ground with a heating coefficient of performance (COP) that theoretically
could be as high as 14 (i.e., it delivers 14 units of heat for 1 unit of electric energy used). Present heat
pumps have COPs closer to 3, but, in theory, they could be as high as 14 for the inside and outside
temperatures mentioned above [Carnahan et al., 1975]. This method would use less than one-seventh
of the energy of our original natural gas furnace. It is a common misconception to believe that there
isn’t much room to improve the efficiency of an “efficient” natural gas furnace.

27. In the 1975 American Physical Society (APS) Energy Efficiency Study this was termed “first-law efficiency” [Carnahan et al., 1975].

28. In the 1975 APS Energy Efficiency Study this was termed, “second-law efficiency” [Carnahan et al., 1975].

29. To actually determine the efficiency of either of the houses one must first determine the minimum primary energy that must be supplied
to heat them—this is a small number, possibly even zero as occupants and sunlight may be sufficient to heat homes.
American Physical Society • September 2008 ENERGY FUTURE: Think Efficiency | 73

Endnote 2. Electricity, Primary Energy and Site Energy

Most of our energy comes from fossil fuels—oil, natural gas and coal. These are primary energy
sources, as are nuclear, hydro, biomass and a variety of renewable sources including wind and solar. A
few primary energy sources, most notably natural gas, are delivered directly to buildings and used on
site. Other forms of primary energy, such as nuclear and hydro, are not delivered directly to buildings,
but instead are used off site for generating electricity. In the United States, energy content of primary
sources is generally measured in British thermal units (Btu), the amount of heat that is required to raise
the temperature of one pound of water by one degree Fahrenheit. One quad is equal to one quadrillion or
1015 Btu.30 One Btu is equal to 1054 joules (J), where the joule is the metric unit for energy.

Electricity is a secondary form of energy. It cannot be gathered, mined or pumped from the ground
but instead must be produced from primary energy sources. Most of our electric energy is generated from
heat produced from coal, natural gas or nuclear energy with an average efficiency of 34 percent—that is,
roughly two-thirds of the primary energy used is lost to waste heat, and only one-third is converted into
electric energy.31 Moreover, 6-7 percent of the electric energy generated is used at the generating plant
or lost in transmission. As a result, only 31 percent of the primary energy consumed is delivered for end
use; the remaining 69 percent is lost as waste heat.

Electric energy is measured in kilowatt-hours (kWh) or their common multiples.32 The watt is a
measure of power, the rate of use (or production) of energy. One watt is equal to one joule per second.
One kWh = 3,600,000 J, the amount of energy used for the length of time of one hour at a rate of 1000 W.
This is also equal to 3,416 Btu.

Though electricity is produced with much inefficiency, it is also a far more useful form of energy
than heat—it powers motors, advanced lighting, computers, and a host of other devices that require work
rather than heat energy. Hence electric energy is vital for buildings. But it is important to recognize that
it comes with a particularly high cost in terms of primary energy and associated greenhouse gas emission
(GHG).

Seventy-two percent of all electric energy in the United States is consumed by the commercial
and residential sectors, with industry and, to a very small degree, transportation, using the rest. Figure
24 shows the flow of primary energy into the electric power sector, and the distribution of electricity
generated to the commercial, residential, industrial and transportation end-use sectors.
The heavy reliance of buildings on electric energy combined with the large losses in generation and
distribution of electricity complicate the process of tracing building energy consumption back to the
primary energy sources. It is far easier simply to total up monthly energy bills and calculate the energy
used at the building itself—the so-called site energy.33 From Figure 24 we see that buildings use 2,646
TWh of electric energy, corresponding to 9.0 quads of site energy due to electricity. But the source
energy34 or primary energy consumed off-site to provide this electric energy is 28.6 quads. And, since
each primary energy source has a different carbon emission factor35, an even more detailed accounting
of primary energy is required to determine greenhouse gas emission associated with building energy
consumption.

30. One quad is also equal to 1.054 exajoules, or the amount of energy contained in 170 million barrels of oil.

31. Newer combined-cycle natural gas plants have much higher efficiencies but do not make up a significant fraction of the nation’s electricity
generating capacity.

32. These include MWh (megawatt-hour = 1 million kWh), GWh (gigawatt-hour = 1 billion kWh), and TWh (terawatt-hour = 1 trillion kWh).
74 |  ENERGY FUTURE: Think Efficiency American Physical Society • September 2008

The flow of electric energy by sector

2006 flows of primary energy into the four end-use sectors (Commercial, Residential, Industrial,
and Transportation) by way of the Electric Power Sector. The electric power sector took in
39.7 quads of primary energy and produced 3,900 TWh of electric energy, 3,650 of which were
delivered to end-use sectors and 250 TWh either used internally or lost in transmission.
The overall efficiency of delivering energy to end-use sectors from primary energy was 31%.

35.5% Commercial
Coal
39.7 quads 3,900 TWh

51.6%
20.7%
3,650 37.0%
Nuclear 15.8% Residential
TWh
9.7% Sales

1.6%
Natural gas 250 TWh
0.6% T&D Loss 27.3%
Industrial
Renewables Coal

Petroleum 0.2%
Other Waste heat Transportation
26.3 quads

The primary energy used by buildings directly as fuels (i.e., delivered to buildings) and indirectly
through electricity is listed in Table 2. Buildings use a total of 10.3 quads of primary sources on site
for fuel, mostly natural gas and petroleum (specifically, home heating oil). Nearly three times as
much primary energy, 28.6 quads, is used by buildings indirectly in the form of electricity, which
brings total primary energy consumption by buildings to 38.9 quads. The last column of the table lists
the associated greenhouse gas emissions in millions of metric tons (megatons) CO2 (MtCO2). Clearly,
coal used for generating electricity—nearly 15 quads—is the dominant source of GHG emission
associated with buildings.

The large disparity between electric site and source energy leads to considerable confusion when
reporting building energy consumption. In this report we use Btu if and only if we are referring to
primary energy and kWh when referring to end-use electric energy. Any exception to this convention
is made explicit in the text.

It is also clear that site energy, while relatively easy to calculate and of some use in comparing
buildings with the same fuel mix, is not a useful concept in determining either GHG emission or energy
security. In general, to have a positive national impact on GHG emission and energy security, the goal
of energy efficiency must be to reduce primary energy consumption, not site energy consumption.
33. For individual buildings, the site energy intensity in Btu/sf is found by adding up the annual purchased energy in Btu and dividing by
the gross square footage of the building. For this calculation, 1 kWh of electric energy is equivalent to 3,416 Btu, ignoring any losses as-
sociated with generating and delivering electric energy.

34. For individual buildings, the source energy intensity in Btu/sf is obtained similarly to the site energy intensity, but by assigning 10,800
Btu of primary energy to each kWh of electric energy, thereby accounting for the 69% average losses in the electric power sector.

35. Carbon emission factors provide the mass of carbon emitted per Btu of energy released.
American Physical Society • September 2008 ENERGY FUTURE: Think Efficiency | 75

Table 2

Primary use of energy in the U.S.

Primary energy use in residential and commercial sectors in quads (1015 Btu), both direct
use of fuels and indirect (i.e., used to generate electric energy supplied). All energy is totaled
for the two sectors and estimated associated greenhouse gas (GHG) emissions calculated in
millions of metric tons carbon.

Residential Commercial Total Est. Total

Energy Source Fuels Electricity Fuels Electricity Energy GHG

(quads) (quads) (quads) (quads) (quads) (MtCO2)
Petroleum 1.4 0.2 0.7 0.2 2.6 190
Natural Gas 4.5 2.3 3.0 2.2 12.1 640
Coal 0.0 7.6 0.1 7.3 14.9 1,460
Renewables 0.5 1.4 0.1 1.4 3.4 0
Nuclear 0.0 3.0 0.0 2.9 5.9 0
Total 6.4 14.6 3.9 14.0 38.9 2,290

Source: Primary energy data from Energy Information Administration 2007 Annual Energy Review; GHG figures are only
calculated estimates.

Endnote 3. Conservation Supply Curves

A useful approach to determine the cost and benefit of energy-efficient measures is through the
use of “conservation supply curves,” which provide estimates of technical-economic potential energy
savings. These are energy savings that can be achieved at a cost lower than the cost of the energy
supply. The curves indicate how much energy saving can be “purchased” for a given cost.

Our examination of supply curves is based on a recent study by Brown, Borgeson, and Koomey
(BBK) that addresses potential energy savings in 2030 [Brown, Borgeson, and Koomey, 2008].
BBK developed separate conservation supply curves for electric and gas end uses for residential and
commercial buildings for the period 2010 to 2030.

Their residential supply curve for electricity use is shown in Figure 25. Also shown is the average
cost of residential electricity (red dashed line at 9.4 cents/kWh). Each bar on the graph has a height
that indicates the cost/kWh of saved energy and a width that represents the total savings in 2030 for
all the cost-effective efficiency measures analyzed in a particular category. The measures chosen were
those analyzed in the report Clean Energy Futures (CEF) [Brown et al., 2001]. The authors updated
CEF using new forecasts of energy use by end use from the Energy Information Administration’s 2007
Annual Energy Outlook. The cost of conserved energy (CCE) is calculated as the present value of the
savings-weighted average for all the measures in that end-use category, using a real discount rate of 7
percent. All savings and costs are expressed in 2007 dollars.

Consider, for instance, lighting efficiency measures, shown as the second bar in Figure 25. An
investment of $2.0 billion reduces 2030 electric consumption by an estimated 169 billion kWh, a
reduction that corresponds to a 1.2 cents/kWh cost of saved energy. The present value of the saved
energy, at an average residential retail rate of 9.4 cents/kWh, is $14.1 billion dollars. This is the kind
of savings that occurs at “net negative cost” since the economic value of the energy saved exceeds
the investment—in this case, considerably. All of the measures shown in Figure 25 have net negative
costs, as each falls below the average price of retail electricity, 9.4 cents/kWh—shown as the horizontal
76 |  ENERGY FUTURE: Think Efficiency American Physical Society • September 2008

Figure 25
Residential electric savings potential for year 2030
Conservation supply curve for electric energy-efficiency improvements in the residential sector. For each
measure considered, (the energy savings is achieved at a cost per kWh less than the average residential retail
price of 9.4 cents/kWh, shown as the horizontal red dashed line.
10
9.4¢ / KWh
Cost of conserved energy (cents/kWh)

9
Freezer
8
Savings potential Dishwasher
7 Space cooling
of 572 TWh Personal computers
6 30% of reference case Refrigeration
Furnace fans
5 Space heater
4 Clothes washer
3 Water
Other uses heating
2
Lighting
Color TV
1
0
0 100 200 300 400 500 600
2030 Energy savings (TWh) 572 TWh

red line in the figure.36 If all of the measures shown in Figure 25 were deployed, they would save 572
billion kWh in 2030 for an investment of $5.2 billion. The value of the saved energy in 2030 at current
prices is $54 billion.

Each of the four supply curves estimated by BBK (of which only one, for residential electricity, is
shown here) indicate that the forecasted building-sector consumption in 2030 can be reduced by about
30% at a cost less than current retail energy prices. These results are similar to results of conservation
supply curves that have been developed over many years. Typically, the cost-effective potential is on
the order of 10 to 15% in the very near term (five years) because of low turnover, 25 to 30% in the
intermediate term (10–20 years), and much higher (depending on the development and experience with
new technologies) in the longer term [Nadel, Shipley and Elliott, 2004].

This approach, like all others, is a simplification. The estimates of technical-economic potentials are
undoubtedly low because they ignore any technologies for which market experience is limited, fail to
consider system integration (e.g., integrated design of lighting, windows, and air conditioning), do not
account for future advances in technology, ignore the economic benefits of reductions in energy prices
due to demand reductions and exclude a variety of other factors.

It is also important to note that there can be a large difference between the estimates of cost-effective
energy savings and those that can be expected in the real world. Because the conservation supply curve
underestimates the technical-economic potential over a 20-year time horizon, we believe it is possible
to achieve these levels even in the face of market barriers. This cannot happen, however, without strong
energy efficiency policies.37

This 30 percent savings will not be achieved by market forces alone. Significant policy tools and
incentives will be required to achieve these results. Yet the committee is skeptical that any reasonable
policies will achieve the 100 percent deployment of cost-effective technologies assumed in this study.
On the other hand, experience shows that some technological advances will occur between now and
2030 and will provide even further efficiency improvements than those included in the BBK study.

36. For those familiar with the McKinsey conservation supply curve—these “negative cost” measures correspond to the left-hand side of
the McKinsey curve.
American Physical Society • September 2008 ENERGY FUTURE: Think Efficiency | 77

The committee believes these two factors are offsetting, so that the 30 percent savings is indeed
achievable.

Historically, three policies have had especially important impacts on energy use in buildings in
the United States—appliance efficiency standards, building energy codes and utility demand-side
management. These are discussed briefly in the text. Here we provide more information.

Endnote 4. Policies That Work

Appliance Standards38
Efficiency standards require products such as refrigerators, electric motors and air conditioners to
meet specific energy requirements. Minimum efficiency standards apply to new equipment sold in the
United States. Consumers still can choose from a range of efficient products with desired attributes
and features.
A History of Federal Support
Minimum efficiency standards for appliances and other equipment were adopted by the federal
government in order to address market failures, replace a patchwork of state standards, save consumers
money and reduce energy use and peak electrical demand. They were first adopted by states, many of
which continue to utilize standards to the extent permitted by federal law. In California, by far the most
active state in setting appliance efficiency standards, the initiative to upgrade and extend standards has
been a hallmark of both Democratic and Republican governors’ programs.

• In 1986, appliance manufacturers and energy efficiency supporters agreed to support uniform
national standards on an array of products. In 1987, President Reagan signed the National
Appliance Energy Conservation Act (NAECA). Standards for fluorescent lamp ballasts were
added by Congress in 1988, and in 1992, President Bush signed the Energy Policy Act that
included efficiency standards for certain types of lamps, electric motors and commercial
heating and cooling equipment.

• In 1989 and 1991, the elder Bush administration issued tougher standards for refrigerators,
clothes washers and dryers, and dishwashers, and began work on several additional standards,
laying the groundwork for the Clinton administration to set new standards for refrigerators,
room air conditioners, ballasts, clothes washers, water heaters and residential central air
conditioners and heat pumps. (The George W. Bush administration reaffirmed the Clinton
clothes washer and water heater standards but tried to weaken the new air conditioner standard
to a seasonal energy efficiency ratio (SEER) of 12, down from SEER 13. That attempt was
overturned in court.)

Efficiency Standards Overcome Market Failures

Minimum-efficiency standards are needed to overcome market failures that restrict the use of
more efficient products. Among these failures are:

• Third-party decision makers (e.g., landlords and builders) who purchase appliances but do not
pay the operating costs of the products they purchase;

• Panic purchases that leave little time for consumers to become educated;

37. Based on previous experience, especially that of California, which has had strong policies to promote energy efficiency in buildings, it
has in fact been possible to achieve the technical-economic potential over a 20-year period, thus attaining much lower electricity growth
than had been forecast.

38. Source: Fact sheet from the American Council for an Energy Efficient Economy, at [Link]
78 |  ENERGY FUTURE: Think Efficiency American Physical Society • September 2008

• Inadequate and misleading information about the relative energy performance of products;
and

• High first costs for efficient equipment due to small production quantities and the fact that
manufacturers frequently combine efficiency features with extra non-energy features in expen-
sive trade-up models.

Energy Efficiency Standards Provide Substantial Public Benefits

Standards enacted to date are having a significant impact on U.S. energy use while saving
consumers and businesses billions of dollars. Appliance standards rank with automobile fuel economy
standards as the two most effective federal energy-saving policies.

• In 2000, according to analyses by the U.S. Department of Energy and the American Coun-
cil for an Energy Efficient Economy (ACEEE), standards reduced U.S. electricity use by
approximately 88 billion kWh and reduced U.S. total energy use by approximately 1,200
trillion Btu. These savings are 2.5% and 1.3% of U.S. electricity and energy use in 2000,
respectively.

• In 2000, standards reduced peak generating needs by approximately 21,000 MW — equiv-

alent to displacing seventy 300 MW power plants. Without those savings, the shortages
during the electricity crisis in California in 2000 and 2001 would have been significantly
worse than they were.

• Over the 1990–2000 period, standards have reduced consumer energy bills by approxi-
mately $50 billion. Under standards, equipment prices have risen modestly, but estimates
by the Lawrence Berkeley National Laboratory and ACEEE indicate that the benefits are
more than three times the costs on a net present value basis.

• As old appliances and equipment wear out and are replaced, savings from existing stan-
dards will steadily grow. By 2010, savings will total more than 250 billion kWh (6.5% of
projected electricity use) and reduce peak demand by approximately 66,000 MW (a 7.6%
reduction). Over 1990–2030, consumers and businesses are projected to save approximate-
ly $186 billion (1997 dollars) from standards already adopted.

• To meet standards, manufacturers often make investments in improving products, but fiscal
impacts on manufacturers are generally modest. For example, in its 1990 Annual Report,
Mor-Flo, a major water heater manufacturer, noted that since NAECA: (1) “we no lon-
ger have to produce models to address the varying state energy efficiency standards;” (2)
“price increases on ... minimum standard models have more than offset the corresponding
cost increases resulting in an improved gross profit margin;” and (3) since the standards
took effect, “the Company has been selling a larger number of ‘step-up’ models.”

Building Energy Codes

Energy codes have a large influence on energy efficiency in states where they are considered
important. In California, energy codes adopted in 1975 have resulted in energy savings of more than
$30 billion, or more than $2,000 per household. To illustrate how effective the codes have been, the
California Energy Commission compiled data on the energy needed to cool a new home, and the
figure decreased by two-thirds from 1970 to 2006. That decrease came despite the fact that the new
homes in 2006 were about 50 percent bigger than the 1970 homes.

California’s code development process shows the value of continuous updating. The state’s
energy code was revised in 2002, 2005, and 2008, and each revision cut energy use by another 10 to
15 percent.
American Physical Society • September 2008 ENERGY FUTURE: Think Efficiency | 79

Unlike California, most states don’t develop their own energy codes but instead follow national model
codes established by the International Codes Council and the American Society of Heating, Refrigerating
and Air-Conditioning Engineers, both nonprofit organizations. ICC and ASHRAE are not mandated to
set strict energy efficient standards, and they have routinely produced weaker model standards based on
consensus rather than on rigorous evaluation of the cost effectiveness of more stringent standards.

There is little federal presence in this area. The Energy Policy and Conservation Act of 1975 required
DOE to set a national code, but the act was repealed in 1981. The only direct federal influence on
codes is for manufactured housing, where the Department of Housing and Urban Development (HUD)
established a national code decades ago. That code hasn’t been revised since 1994 and is significantly
weaker than the International Energy Conservation Codes. DOE has some code modeling programs, but
the annual budget is less than $10 million, and the department has not taken a leading role in supporting
major improvements in efficiency.

Codes are enforced at the local level, not by the organizations that develop and adopt the codes, and
this split responsibility is a cause of inadequate enforcement. The effectiveness of code enforcement has
not been measured very often or very rigorously, but the available evidence suggests that states that put
a priority on effective enforcement through adequate staffing and training programs get results within 10
percent of the expected energy consumption. States without such programs do considerably worse.

The California Energy Commission has estimated that a good nationwide program to enhance energy
code enforcement would cost about $50 million annually. If it improved energy performance in new
construction by only 10 percent, it would save about $300 million a year.

Utility Demand-Side Management Directed at Customer Energy Efficiency

Demand-side management (DSM) programs, as noted in the body of the report, assist customers
in becoming more energy efficient through education programs, efficient appliance purchase rebates,
weatherization programs and other methods typically financed by utilities. DSM programs involving
customer energy efficiency improvements have played two important roles. First, they have provided
subsidies for energy efficiency purchases by consumers, reducing growth of kWh in the short term.
Second, utility DSM programs have transformed markets by bringing energy-efficient products into
widespread use.

Analysis of specific DSM programs has shown benefits greater than costs. Such analysis would
ideally consider the amount spent on energy-efficiency programs in a given year along with the total
energy savings (in that and all future years) resulting from the investment. Such data are not readily
available. Instead, the Energy Information Administration asks utilities to annually report the amount
spent on energy efficiency programs and load management programs,39 along with the estimated energy
savings achieved in that same year due to these and all previous such investments.

Table 3 shows these data for 1995 through 2006. The direct expenditure (in millions of dollars) on
energy efficiency programs is shown in column 2, the direct expenditure on load management programs
in column 3 and the indirect expenses for administering both programs in column 4. To obtain the total
spent on just energy efficiency, we distribute the indirect costs to energy efficiency and load management
in the same proportion as their direct costs and arrive at the total spent on energy efficiency in column 5.
The last column gives the total electric energy saved in that year due to these and all previous investments.
For the 12 consecutive years, $13.8 billion was spent on DSM energy efficiency programs resulting in
roughly 650 TWh of saved electric energy.

Endnote 5. Integrated Design

Because designing, constructing and operating very-low-energy commercial buildings is so difficult,

39. Discussions with Energy Information Administration staff

80 |  ENERGY FUTURE: Think Efficiency American Physical Society • September 2008

they are rare. The two major

Table 3 databases that collect information
Demand-side management programs on high-performance buildings
Demand-side management program direct and indirect costs (in millions of dollars) both show a paucity of very-low-
Indirect costs are overhead costs not attributable to specific programs. Total energy energy buildings. One database
efficiency cost is the sum of direct EE plus a proportional amount of the indirect costs 40 is maintained by DOE (http://
[Link]/buildings/
Energy Load Indirect Total energy Energy savings database/[Link]) and the other
efficiency mgmt. cost efficiency (GWh) is compiled by the New Buildings
1995 $1,409 $569 $419 $1,701 55,328 Institute on buildings that meet
1996 $1,052 $572 $279 $1,232 59,853 the Leadership on Energy and
1997 $892 $455 $289 $1,084 55,453 Environmental Design (LEED)
1998 $766 $467 $188 $883 48,775 standard ([Link]
1999 $820 $431 $173 $934 49,691 org/).41 Together they contain
2000 $939 $446 $181 $1,061 52,827 energy consumption data for 159
2001 $1,098 $358 $175 $1,229 52,946 commercial buildings, which are
2002 $1,007 $414 $205 $1,152 52,285 intended to represent the most
2003 $807 $352 $138 $903 48,245 energy-efficient commercial
2004 $910 $515 $132 $995 52,662
buildings in the United States. In
these databases we find only 17
2005 $1,169 $626 $127 1,252 58,891
U.S. commercial buildings with
2006 $1,258 $666 $127 $1,342 62,951
measured annual site energy
TOTAL: $12,128 $5,896 $2,429 $13,768 649,907 intensity that is 70 percent lower
Source: Energy Information Administration
than the national average for
all commercial buildings. And
these very-low-energy buildings
are usually the smaller ones—
together they contain less than 4% of the gross square footage of the 159 buildings. Even more
disconcerting is the fact that the average annual site energy intensity (on a per square foot basis) for
these 159 relatively new, “green” buildings is actually higher than the average for the entire U.S.
commercial building stock—calculated either as site or primary energy..42

Although it is impossible to extrapolate from these limited databases to all new commercial
building construction, it is clear that the number of very-low-energy commercial buildings constructed
annually in the United States is small. On the positive side, the 17 buildings that have achieved the
70 percent reduced energy level demonstrate that it can be done.

Why are so few such low-energy commercial buildings being constructed?

A “commercial building is a complex system, with the energy use and performance of any one
part of the system affecting the energy use of the building as a whole through a complex cascade
of interactions. However, the typical design process for commercial buildings is a linear, sequential
process that precludes the analysis and design of the buildings as an integrated system. In order to
achieve deep savings in energy use, an integrated and iterative design process, involving all members
of the design team, is required” [Harvey, 2006].

The integrated process may be defined as one in which “all of the design variables that affect one
another are considered together and resolved in an optimal fashion” [Lewis, 2004]. The sequence
40. Source: Energy Information Administration, Form EIA-861, “Annual Electric Power Industry Report.”

41. The data for the 121 LEED buildings were made available to us by Cathy Turner, New Buildings Institute.

42. The EIA’s 2003 Commercial Building Energy Consumption [Link] the average annual site energy intensity for nearly 4.7
million non-vacant buildings to be 95 kBtu/sf, which is 23 percent lower than the average for these 159 buildings. Primary energy data
are available for 132 of the 159 buildings. Average primary or source energy consumption for this subset is 38% higher than the CBECS
average of 198 kBtu/sf.
American Physical Society • September 2008 ENERGY FUTURE: Think Efficiency | 81

of steps that is typically followed today often leads to solutions that are far from optimal. For example,
HVAC capacity and equipment are often decided before the major contributors to the internal loads of
a building are known.

Significant interactions among all design elements of a building affect heating and cooling loads (e.g.,
window size, placement, and thermal characteristics; window shading types and placement; lighting
locations, efficacy and local controls; building orientation; number and wattage of plug loads; and the
volume of outside air that is circulated into a building).

All of these elements need to be considered in light of advanced technology options (e.g., on-
site generation, passive ventilation, thermal mass with night ventilation, chilled ceiling displacement
ventilation, dehumidification and daylighting). Control strategies and operating conditions for all of the
equipment in the building strongly affect the effectiveness of the design and technology choices for the
building.

Finally, all of these complex design and engineering issues must themselves be integrated with
decisions on structural issues, space planning, site context, materials selection and other issues, all
within the context of tight budgets and schedules.

To address these interactions among the different components of a building, integrated design and
operation requires cooperation among the major decision makers in a building project—architects,
engineers, and builders—to evaluate the projected energy consumption for a variety of designs. Building
professionals must also enjoy a comfort level in using results of computer tools to underpin important
design decisions. Software that is understandable to everyone involved is needed, so that the group’s
collective knowledge is codified and used as different problems and solutions are addressed in the design,
construction, and eventually the operation of the building.

The need goes beyond the design process. Even the best-designed buildings, with well-thought-out
integrated systems, can suffer in their construction by contractors who lack the skills and experience
to implement the details faithfully. And facility managers may not know how to operate a new system
properly. A $100 home appliance comes with a setup and operating manual; many buildings do not.

Just as any complex machine gets out of tune with use, commercial buildings decline in performance
if problems are not corrected. A recent study of U.S. commercial buildings found that HVAC equipment
that is either faulty or not operated properly accounts for between 2 and 11 percent of energy consumption
[TIAX, 2005]. Performance would be improved by an advanced diagnostic and control system, running
alongside a real-time building energy simulator, with sensors to collect operating data, identify problems
as they occur and recommend adjustments.
82 |  ENERGY FUTURE: Think Efficiency American Physical Society • September 2008

Figure 26 is a schematic diagram of the technologies that could enable integrated operation of a
commercial building of the future.

Figure 26
A commercial building with integrated
systems design and operations
This diagram of a large commercial building shows various aspects and systems whose
interactions are important for optimum building operation and minimum energy consumption.

Oversized Fuel cell Roof

cooling PV panels
tower

Electrochromatic glazing Weather station for load

integrated with thermal prediction and control
and daylighting systems
Dimming ballasts with
efficient fixtures
VAV boxes with duct
Automated blinds pressure reset
Electric load optimization
with real time pricing
among multiple buildings

Automated blinds Occupancy controls

air quality sensors
Awnings

Light shelves

Intergrated controls with real

Efficient time modeling, fault diagnostics
chiller with and optimization
heat recovery Efficient boiler
with heat recovery

Detailed area

Source: Lawrence Berkeley National Lab

American Physical Society • September 2008 ENERGY FUTURE: Think Efficiency | 83

Chapter 3 References
• 2007 Buildings Energy Data Book, prepared for the Buildings Technologies Program and
Office of Planning, Budget, and Analysis, Energy Efficiency and Renewable Energy, U.S.
Department of Energy by D&R International, Ltd., September 2007.

• American Council for an Energy Efficient Economy (ACEEE), [Link]

[Link] (2008).

• G. J. Brohard et al., Advanced Customer Technology Test for Maximum Energy Efficiency
(ACT2) Project: The Final Report, Pacific Gas & Electric, 1997.

• M. A. Brown, M. D. Levine, W. Short and J. G. Koomey, “Scenarios for a Clean Energy

Future,” Energy Policy 29, 1179-1196 (2001).

• R. Brown, S. Borgeson and J. Koomey, Building-Sector Energy Efficiency Potential Based on

the Clean Energy Futures Study, 2008, to be published.

• C40 Cities Climate Leadership Group, [Link]

berlin_efficiency.jsp, 2008.

• W. Carnahan, K. W. Ford, A. Prosperetti, G. I. Rochlin, A. Rosenfeld, M. Ross, J. Rothberg, G.

Seidel and R. H. Socolow, eds., Technical Aspects of the More Efficient Utilization of Energy,
Conference Series, vol. 25, American Institute of Physics, New York (1975).

• R. Cavanagh, Energy Efficiency in Buildings and Equipment: Remedies for Pervasive Market
Failures, National Commission on Energy Policy, 2004.

• J. Christian, Affordable and Demand Responsive Net Zero House, draft report, private
communication (2008).

• Committee on Benefits of DOE R&D on Energy Efficiency and Fossil Energy, Energy Research
at DOE: Was It Worth It? Energy Efficiency and Fossil Energy Research 1978 to 2000 National
Academy Press, Washington, DC (2001).

• L. R. Glicksman, “Energy Efficiency in the Built Environment,” Physics Today, July 2008, p.
35.

• D. B. Goldstein, Saving Energy, Growing Jobs, Bay Tree Publishing, Berkeley, CA (2007).

• D. B. Goldstein and M. G. Hoffman, “Forecasting an Increasing Role for Energy Efficiency in

Meeting Global Climate Goals,” Proceedings of the 2004 Summer Study on Energy Efficiency
in Buildings, American Council for an Energy Efficient Economy, Washington, DC, August
2004.

• L. D. Danny Harvey, ed., A Handbook on Low-Energy Buildings and District-Energy Systems:

Fundamentals, Techniques and Examples, Earthscan Publications Ltd., London, 2006.

• International Energy Agency, Mind the Gap: Quantifying Principal-Agent Problems in Energy
Efficiency, IEA, Paris, 2007.

• I. Johnstone, “Energy and Mass Flows of Housing: Estimating Mortality,” Building and
Environment, v.36, pp.43-51 (2001).
84 |  ENERGY FUTURE: Think Efficiency American Physical Society • September 2008

• M. Lewis, “Integrated Design for Sustainable Buildings, Building for the Future, ASHRAE
Journal, Supplement, p. S22 (September 2004).

• V. Loftness, “Improving Building Energy Efficiency in the U.S: Technologies and Policies
for 2010 to 2050,” in proceedings of the workshop, The 10-50 Solution: Technologies
and Policies for a Low-Carbon Future, sponsored by the Pew Center on Global Climate
Change and the National Commission on Energy Policy (2004).

• C. Marnay, G. Venkataramanan, M. Stadler, A. S. Siddiqui, R. Firestone and B. Chandran,

“Optimal Technology Selection and Operation of Commercial-Building Microgrids,”
IEEE Transactions on Power Systems (2007), and references therein.

• J. McMahon, “Buildings: Energy Efficiency Options,” presentation at the Haas School,

University of California, Berkeley, Nov. 27, 2007.

• A. Meier, Lawrence Berkeley National Laboratory, private communication (2008).

• S. Nadel, A. M. Shipley and R. N. Elliott The Technical, Economic, and Achievable

Potential for Energy Efficiency in the United States: A Meta-Analysis of Recent Studies,
American Council for an Energy Efficient Economy, (Washington, DC, 2004).

• S. Nadel et al., Leading the Way: Continued Opportunities for New State Appliance and
Equipment Standards, ACEEE, Washington DC (2006); also see Energy Bill Savings
Estimates as Passed by the Senate, ACEEE, Washington DC (December 14, 2007), posted
at [Link]

• P. Norton et al., Affordable High-Performance Homes: A Cold-Climate Case Study, NREL/

TP-550-31650, National Renewable Energy Laboratory, Golden, CO, April 2005.

• Panel on Policy Implications of Greenhouse Warming, Policy Implications of Greenhouse

Warming: Mitigation, Adaptation, and the Science Base, National Academy Press,
Washington, DC (1992).

• A. Rosenfeld, California Energy Commission, private communication (2008).

• J. L. Barrientos Sacari, U. Bhattacharjee, T. Martinez and J. J. Duffy, “Green Buildings

in Massachusetts: Comparison Between Actual and Predicted Energy Performance,”
Proceedings of the American Solar Energy Society, Cleveland, OH, July 9-13, 2007.

• J. Schnieders, “CEPHEUS—Measurement Results from More than 100 Dwelling Units in

Passive Houses,” posted at [Link]

• A. Sudarshan and J. Sweeney, “Deconstructing the ‘Rosenfeld Curve’,” submitted to

Energy Journal (2008).

• TIAX LLC, Energy Impact of Commercial Building Controls and Performance Diagnostics:
Market Characterization, Energy Impact of Building Faults and Energy Savings Potential,
final report prepared for the U.S. Department of Energy Building Technologies Program,
Washington, DC (November 2005).

• C. Turner and M. Frankel, Energy Performance of LEEP for New Construction Buildings,
Final Report, New Buildings Institute, White Salmon, WA, 2008
American Physical Society • September 2008 ENERGY FUTURE: Think Efficiency | 85

• U.S. Environmental Protection Agency, Energy Star Program, Frequently Asked Questions:
Information on Compact Fluorescent Light Bulbs (CFLs) and Mercury, [Link]
gov/ia/partners/promotions/change_light/downloads/Fact_Sheet_Mercury.pdf, June 2008.

• C. Zimmerman, Wal-Mart, “The Continuing Evolution of Sustainable Facilities at Wal-Mart,”

private communication, (2008).