Energy Information Administration Home Page Welcome to the Energy Information Administration's Energy Efficiency Web Site.  If you need assistance in viewing this page, please call (202) 586-8800

  Home > Energy Users > Energy Efficiency Page > The Commercial Building Sector - Chapter 4

4. Commercial Building Sector


The commercial building sector is extremely difficult to delineate. As generally covered by EIA, and for the purposes of this document, the commercial building sector consists of all activities other than transportation whose principal activities not residential or industrial(16) businesses. The vast majority of energy use associated with this sector occurs in buildings, to maintain the building environment, and provide building-based services. Commercial buildings encompass diverse kinds of structures with equally varied operations. Energy use in the commercial building sector is affected by the physical characteristics of the buildings, age of the buildings, efficiency of the equipment, occupants' energy-related behavior, location, and structural effects. Different principal building activities(17) demand different energy services and various levels of a particular service. The energy used in commercial buildings range from the major energy sources--electricity, natural gas, fuel oil, and district heat--to the minor energy sources: propane, coal, and the renewable sources, wood and solar.(18)

The energy supply disruptions of 1973 and 1979, along with a growing concern for the environment, led to increased awareness of the need to reduce energy consumption. Commercial buildings built in the 1980's were designed for specific climates, were well-insulated, and included more efficient space conditioning and lighting systems. Nevertheless, an expanding economy and increases in energy service demand during the 1980's caused an increase in energy consumption in the commercial building sector. During that period energy demand grew by 1.0 percent a year.(19)

Between 1979 and 1983, the economy was growing at approximately the same rate as total primary energy consumption Figure 4.1). In 1983 total energy consumption and Gross Domestic Purchases diverged, reflecting the growing concern for the environment and the building of more energy-efficient buildings.(20) This gap seems to narrow somewhat in 1989 only to reopen when the economy faced a slowdown before the recession in 1991. Between 1989 and 1992, total site energy consumption remained relatively stable while total primary energy fell slightly, reflecting a possible relative decline in electricity consumption. In 1989 electricity consumption was 48 percent of all commercial site energy consumption, but in 1992, this percentage fell to 45 percent.

Chapter Organization

In this chapter, the major data source, EIA's Commercial Buildings Energy Consumption Survey, is described first. A discussion of site energy consumption in the commercial building sector follows, along with a discussion of the necessity of adjusting energy consumption for weather and vacancy influences. Next, the demand indicators, buildings, floorspace, square-foot hours, number of employees, and gross domestic product for services are described along with the trends in and adjustments to these demand indicators. Five energy-intensity indicators will be presented followed by a discussion of the strengths and limitations of the energy-intensity indicators.

Major Data Sources

Energy Information Administration

Commercial Buildings Energy Consumption Survey (CBECS)

The CBECS is a national representation probability sample of commercial buildings. For the purposes of this survey, a commercial building is defined as one whose principal activity is not residential or industrial. The survey covers all commercial buildings over 1,000 square feet. For each of the roughly 6,000 buildings in the sample, the CBECS collects data on energy-related characteristics of the building through personal interviews with the buildings' owners or managers, and total energy consumption for all end uses from billing data provided by the buildings' energy suppliers.

EIA conducts this national sample survey of commercial buildings and their energy suppliers on a triennial basis. Previous surveys were conducted in 1979, 1983, and 1986 under the name Nonresidential Buildings Energy Consumption Survey (NBECS). In 1989, the survey name was changed to CBECS and for consistency, all the surveys will be referred to CBECS in this chapter. The latest CBECS was in 1992.

Energy Consumption in the Commercial Building Sector


Energy is used in the commercial building sector to provide services such as lighting, space conditioning, ventilation, water heating, refrigeration, powering office equipment, and other uses. The amount of energy used to provide these services depends on the activities taking place in the commercial buildings; for example, health care, offices, food service, and so forth. EIA classifies commercial buildings by the activity occupying the most floorspace in the building. This classification is based on the premise that buildings within the same principal building activity have similar energy consumption patterns and that activity classification is useful to examine differences in energy use among various types of buildings.

In 1992, 27 percent of the 4.8 million commercial buildings in the United States were mercantile buildings used primarily for sale and/or distribution of goods and services. These buildings had 18 percent of the 67.9 billion square feet of commercial floorspace and used 15 percent of the 5.8 quadrillion Btu of site commercial energy to provide energy services. Office buildings accounted for only 16 percent of all buildings and floorspace, but commanded 22 percent of all the commercial site energy used in 1992 (Figure 4.2. and Figure 4.3). (21)

In 1989, the largest proportion of the energy used for energy services in commercial buildings was for space heating (35 percent) followed by lighting (18 percent), water heating (9 percent), office equipment (7 percent), space cooling (5 percent), ventilation (5 percent), cooking (5 percent) and refrigeration (3 percent).(22) Some building activities use more of a particular energy service than other building activities; for example, office buildings used only 30 percent of the total site energy for space heating, but educational buildings used 54 percent of total site energy for space heating.

Additionally, regional differences affect the level of demand for energy services. For example, in 1989 the South Census Region, a very warm region, used 8 percent of its total site energy for space cooling, whereas the Northeast, a cooler region, used only 3 percent of total site energy for space cooling.

Energy Trends

Two intervals of time will be used for trends, similar to those used for the residential sector but following the years of the CBECS. The first is an interval of growth/growth (1986 to 1989) and the second of recession/recovery (1989 to 1992). Total site energy consumption grew by 16 percent in the first interval and by less than 1 percent during the second (Table 4.1 and Figure 4.4).

During the growth/growth interval, total site energy consumption increased for each building type except for food and health care, but total site consumption would have increased more if vacancy rates had not increased. During those years, new commercial buildings were constructed at a rapid rate, and the total floorspace of vacant buildings doubled, increasing from 2,090 million square feet in 1986 to 4,161 million square feet in 1989.

During the recession/recovery interval, changes in total site energy consumption varied by building activity (Table 4.1). Floorspace for vacant buildings continued to grow, but only by 6 percent. The decline in the rate of growth of vacant space was attributed to the decline in growth of commercial building construction and leasing costs.

Weather affects energy consumption in commercial buildings. During the intervals studied, the winters were warmer and the summers were hotter than the 30-year average. If the weather had been closer to the average, total site energy consumption would not have increased as much during the growth/growth interval, and would have grown instead of staying flat during the recession/recovery interval.(23)

Total site energy consumption estimates can be adjusted not only for the weather effects, but also for vacancy effects. Adjusting estimates to omit vacant buildings, which use little energy, changed the trend in total site energy consumption during the intervals under study. During the growth/growth interval, the increase in adjusted consumption was less pronounced than the unadjusted trend, and the growth in total site energy consumption during the recession/recovery interval was even larger (Table 4.1 and Figure 4.4).(24)

Table 4.1. U.S. Commercial Total Site Energy Consumption by Principal Building Activity, 1986, 1989, and 1992  (Million Btu)

Principal Building Activity Unadjusted Weather Adjusted Weather and Vacancy Adjusted
1986 1989 1992 1986 1989 1992 1986 1989 1992
All Buildings 4,977 5,788 5,803 5,057 5,725 5,931 4,757 5,317 5,664
Principal Building Activity

Public Assembly



Health Care

































































































-- = Data not available.

Sources: Energy Information Administration, Office of Energy Markets and End Use, 1986, 1989, 1992 Commercial Buildings Energy Consumption Surveys, Public-Use Data Files.

Eliminating effects of extreme weather reduces the peaks and fills the valleys. By eliminating vacant building energy use, periods of high vacancy rates (e.g., 1989) are corrected to reflect occupied energy use. For almost all principal building activities, these adjustments reduced the magnitude of the decrease and augmented the increases in site energy consumption.

Demand Indicators

In the commercial building sector, some of the demand indicators for energy services are: number of buildings, operating hours, number of workers in a building, and size of a building by floorspace. The greater the floorspace (measured in square feet), number of buildings, and operating hours of buildings, the higher the energy required for heating, cooling, lighting, and ventilation and other end uses. Thus, the increases and decreases in unadjusted total site energy consumption shown in Table 4.1 and Figure 4.4, for instance, reflect commensurate increases and decreases in many of the commercial demand indicators (Figure 4.5).

Trends in Demand Indicators

Number of Buildings

The number of commercial buildings increased by 9 percent during the growth/growth interval and by 6 percent during the recession/recovery interval (Figure 4.6). During both intervals construction of new buildings decreased dramatically, dropping by 25 percent over the growth/growth interval and by 40 percent over the recession/recovery interval.(25)

Total Floorspace

Patterns of growth in total floorspace were similar to those for number of buildings during both time intervals. Increases in total floorspace for new construction also followed the growth in the number of buildings over the growth years; however, during the recession/recovery interval, new total floorspace decreased by 23 percent although new construction fell by approximately 40 percent. In 1986, newly constructed buildings averaged 15.8 thousand square feet per building. This average did not change in 1989, but in 1992, the average square footage of a newly constructed building was 19.6 thousand square feet.

Square-Foot Hours

Square-foot hours are a surrogate measure of economic activity.(26) Square-foot hours grew rapidly during the growth/growth interval (22 percent). The increase in this indicator may be attributed to the growth of retail sales in mercantile and service buildings. Retail sales grew by almost 10 percent (constant 1987 dollars) during this time.(27) The rate of growth in square-foot hours as well as the rate of growth in retail sales slowed during the recession/recovery interval, resulting in an increase of only 2 percent for retail sales and 3 percent for square-foot hours. The percent of all commercial buildings open 48 hours or less per week increased from 40 percent in 1989 to 48 percent in 1992. Conversely, the percent of all commercial buildings open 85 hours or more per week decreased from 20 percent in 1989 to 17 percent in 1992. The reduction in operating hours follows the economic slowdown the country as a whole was facing during those years.

Number of Employees

The total number of employees fell during the growth years and was flat during the recession/recovery interval.

During the growth years, the number of buildings with fewer than 10 employees grew by 7 percent, whereas the number of buildings with 10 to 19 employees fell by 4 percent. The commercial building sector, perhaps following the manufacturing sector, may have downsized. Investigation into changes in the number of buildings in each principal building activity type may be warranted since some principal building activity types may be more apt to substitute equipment for employees than others (e.g., increased use of computers in office buildings to increase productivity).

Box 4.1

Demand Indicators in the Commercial Building Sector

  • Number of Buildings
  • Floorspace
  • Square-Foot-Hours
  • Number of Employees
  • Gross Domestic Product for
    Services (Constant 1987 Dollar)

Gross Domestic Product (Services)

This estimate of economic activity appears to run counter to the results of changes in square-foot hours, a surrogate to economic activity. During the growth/growth interval, the growth in Gross Domestic Product (GDP) for services was far smaller than the growth in square-foot hours. One possible explanation is that commercial businesses may have kept their doors open longer to make the same profit that they made in fewer hours in better economic times.

Demand-Indicator Adjustments

Demand indicators, with the exception of GDP for services, can be adjusted by removing vacant buildings from the estimates in the same way as the consumption estimates were adjusted. The effects of these adjustments are depicted in Figure 4.6.

Energy-Intensity Indicators for the Commercial Building Sector

Energy-intensity indicators are used to measure the ratio of energy consumption to the demand for services using the demand indicators described in the previous section. The energy-intensity indicators may be applied across the entire sector, or conditionally for a specific building activity, building age, end use, or other limiting characteristics. A commonly used energy-intensity indicator for the commercial building sector is energy consumption per square foot. However, a variety of other popular indicators exist (see Box 4.2).

Box 4.2

Energy-Intensity Indicators for the
Commercial Building Sector

  • Million Btu per Building
  • Thousand Btu per Square Foot
  • Btu per Square-Foot-Hour
  • Thousand Btu per Employee
  • Thousand Btu per Dollar of Gross Domestic
    Product for Services (Constant 1987 Dollar)

Trends in Energy-Intensity Indicators

During the growth/growth interval, all the energy-intensity indicators--except thousand Btu per square-foot hour(28)--registered increases in intensity, a trend suggesting that energy efficiency decreased.(29) During the recession/recovery interval, all of the energy-intensity indicators decreased, suggesting increases in energy efficiency (Figure 4.7). The results in the two different intervals suggest that influences other than only changes in energy efficiency may be affecting the results. If the time interval chosen were the entire period, 1986 to 1992, the energy-intensity indicators, million Btu per building and thousand Btu per square foot, showed no changes (Table 4.2).

Adjusting demand indicators in the development of the energy-intensity indicators begins to remove a few of those influences that affect energy-intensity that are not due to changes in energy efficiency (Table 4.2 and Figure 4.7). The adjustments seem to put a damping effect on the intensity indicators. The increases in intensity during the growth/growth interval are less pronounced, and the decreases during the recession/recovery interval are not as great as the unadjusted estimates.

Table 4.2. Comparison of Commercial Site Energy-Intensity Indicators, 1986, 1989, and 1992

Energy-Intensity Indicators Units
1986 1989 1992
Unadjusted Weather and Vacancy Adjusted Unadjusted Weather and Vacancy Adjusted Unadjusted Weather and Vacancy Adjusted
Million Btu/Building 1,198 1,300 1,278 1,358 1,207 1,325
Thousand Btu/Square Foot 86 92 92 97 86 93
Btu/Square-Foot-Hour 24 25 23 24 22 23
Thousand Btu/Employee 68 69 82 81 82 83
Thousand Btu/Dollar GDP 2,231 2,132 2,534 2,328 2,379 2,332
Note: Service sector Gross Domestic Product estimates are in 1987 constant dollars.

Sources: Energy Information Administration, Office of Energy Markets and End Use, 1986, 1989, and 1992 Commercial Building Energy Consumption Surveys, Public-Use Data Files. U.S. Bureau of Economic Analysis, National Income and Product Accounts of the United States, volume 2, 1959-1988, and Survey of Current Business, March 1993.

Examination of estimates of the unadjusted and adjusted energy-intensity indicator (thousand Btu per square-foot-hour) across the four Census regions demonstrates regional differences (Figure 4.8). During the interval of growth/growth, the Northeast was the only Census region with increases in the intensity indicator, even after adjustments are made. During the recession/recovery interval, the Northeast experienced the largest decrease in the intensity indicator (-19.7 percent unadjusted and -17.1 percent adjusted). The Northeast was the only Census region to have a negative energy-intensity indicator. Most of the large decrease was in the mercantile and warehouse building activities (Figure 4.9).

In the Northeast, in 1992, the demand for electricity was 38 percent of the demand for all major fuels. In the early 1990's, electric utilities aggressively promoted demand-side management (DSM) programs, especially in the Northeast to either reduce electricity demand at peak times or throughout the year. According to the electric utilities in the Northeast, 55 percent of the commercial buildings participated in some type of DSM. This may have been one of the contributing factors in the reduction of the energy-intensity indicator.

Going one step further is an examination of the estimates for an energy-intensity indicator, thousand Btu per square-foot-hour, by principal building activity in the Northeast. Figure 4.9a shows that lodging, mercantile, health care, and warehouse building types displayed increases in the energy-intensity indicator during the growth/growth interval while other activities posted reductions. These relationships remained even when the estimates were adjusted for weather and vacancy.(30)

Other structural and behavioral effects need to be considered, such as new construction growth and changing business tax laws in addition to the weather, vacancy, regional, and economic effects discussed above. Such consideration is necessary to separate the effects that are related to energy efficiency from the effects that are unrelated. Standardizing could take place by basically keeping a characteristic such as floorspace at a certain level and examining the intensity indicators as if the amount of floorspace had remained the same. However, as more standardizations take place, a greater level of detailed data will be needed.

Strengths and Limitations of the Energy-Intensity Indicators

The CBECS contains sufficient detail that site energy consumption estimates can be adjusted for some of the major influences on changes in energy consumption such as weather and vacancy. The CBECS data, though, are available only every 3 years. However, this limitation may not be important because the measurement of energy-intensity may be feasible only every few years because of the availability of other data and limited resources.

Table 4.3 presents some of the strengths and limitations of the demand indicators that influence the amount of energy consumed in the commercial sector. On needs to remember that the greater the disaggregation of the energy-intensity indicator, e.g., a particular energy service such as space heating, the greater are the data needs and the lower the precision of the data due to sample-size limitations. The particular use of an indicator, such as for policy program evaluation, may dictate the type of indicator and the level of disaggregation needed.

Table 4.3. Strengths and Limitations of the Energy-Intensity Indicators

Energy-Intensity Indicator
Strength of Demand Indicator Used
in the Energy-Intensity Indicator
Limitation of the Demand Indicator Used
in the Energy-Intensity Indicator
Energy per Building
  • Collected by EIA as part of a detailed energy survey
  • Vacancy adjustments are possible
  • Data available every 3 years
  • Data not available for buildings 1,000 square feet or less
Energy per Square Foot
  • Collected by EIA as part of a detailed energy survey
  • Demand for major energy services such as space conditioning is influenced by the amount of floorspace
  • Vacancy adjustments are possible
  • Data available every 3 years
  • Does not completely capture the operational profile of the building, e.g., water heating does not depend on floorspace in the building
  • Data not available for buildings 1,000 square feet or less
Energy per Square-Foot Hour
  • Calculated from EIA CBECS data
  • Surrogate for economic activity
  • Demand for major energy services such as space conditioning is influenced by the amount of floorspace
  • Vacancy adjustments are possible
  • Data available every 3 years
  • Does not completely capture the operational profile of the building, e.g., water heating does not depend on floorspace in the building
  • Data not available for buildings 1,000 square feet or less
Energy per Employee
  • Available annually
  • Capture the operational profile of the building for water heating for certain building activities, e.g., office buildings
  • Not available by detailed characteristics
Energy per GDP Services (1987 Dollars)
  • Measure of economic activity
  • Available annually
  • Not available by detailed characteristics

End Notes            

16 Industrial businesses include manufacturers; agriculture, forestry, and fisheries; mining; and construction. These are excluded from the Commercial Building Sector.

17"Principal Building Activities" is defined in the commercial section of the Glossary.

18In this chapter, commercial site energy consumption includes only electricity, natural gas, fuel oil (distillate and residual), and district heat(hot water and steam).

19See EIA's Energy End-Use Intensities in Commercial Buildings (DOE/EIA-0555(94)/2), pp. 1 and 2.

20Gross Domestic Purchases is a measure of U.S. gross domestic purchases of goods and services including imports and excluding exports. Whereas Gross Domestic Product (GDP) is a measure of U.S. gross domestic production of goods and services. Gross Domestic Product for only services is used later in the chapter. See "Gross Domestic Purchases" and "Gross Domestic Product" in the general terminology section of the Glossary.

21EIA's Commercial Buildings Energy Consumption and Expenditures 1992 (DOE/EIA-0318(92)) presents a profile of energy consumption and expenditures in 1992.

22At the present time, only the 1989 CBECS data have been used to estimate end-use estimates. Engineering estimates were derived for each building in the sample. These estimates were then statistically adjusted to match the total energy consumption for each building. This methodology is fully developed in Energy End-Use Intensities in Commercial Buildings (DOE/EIA-0555(94)/2), September 1994.

23See "Degree-Day Adjusted Estimates" in the commercial section of Appendix A for the methodology and an example.

24See "Occupied Commercial Buildings Site Energy Consumption Adjustment" in the commercial section of Appendix A.

25 Data obtained from EIA's 1986, 1989, and 1992 Commercial Buildings Energy Consumption Surveys. Other undocumented data presented in this chapter were also obtained from the respective CBECS.

26For each principal building activity, the total building square footage is multiplied by the annual operating hours. The square-foot hours are summed over all principal building activities to obtain total square-foot hours.

27U.S. Department of Commerce, Bureau of Economic Analysis, Survey of Current Business, various March issues.

28In the last section it was shown that square-foot hours had increased during the growth/growth interval by approximately 18 to 22 percent depending on whether the indicator was adjusted for weather and vacancy. This rate of increased slowed dramatically over the recessionary/ recovery interval.

29Given the variety of energy-intensity indicators available, caution is warranted in attempting to compare indicators. The magnitude and direction of changes in the energy-intensity indicators are dependent on the choice of demand indicator used in the denominator. Additionally, structural or behavioral effects may intensify or diminish changes in energy-intensity.

30Adjustments are not uniform over the principal building activities since for each of the building activity types energy demand for different energy services and different levels of energy services are not the same.

Back to the Table of Contents     Table of Contents blue_next.gif (2052 bytes)    Transportation Sector


Specific questions on this topic may be directed to:
Stephanie Battles
(Phone: (202) 586-7237)
FAX: (202) 586-0018

Contact Us


File Last Modified: October 17, 1999