GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Plus Icon Minus Icon Picture icon Hamburger Icon Close Icon Sorted
Energy Solutions

Energy Use by Buildings

If we include building-related portions of industrial and transportation sectors, buildings account for a lot more energy than most people calculate

Image 1 of 4
The generally accepted split between different primary energy end-uses. Click the image to enlarge it.
Image Credit: U.S. Department of Energy, Energy Information Administration
The generally accepted split between different primary energy end-uses. Click the image to enlarge it.
Image Credit: U.S. Department of Energy, Energy Information Administration
Historical U.S. primary energy consumption by end-use sector.
Image Credit: U.S. Department of Energy, Energy Information Administration
If the primary energy used for electricity generation is apportioned to the end-use categories, the totals (in blue) result.
Image Credit: U.S. Department of Energy, Energy Information Administration
Transportation energy use in the U.S. by mode - 2010.
Image Credit: Transportation Energy Data Book - Edition 31, U.S. Department of Energy

I’ve long appreciated the adage that you can’t manage what you don’t measure, so I’ve spent a good bit of time looking at numbers — especially relating to energy. (Apologies in advance to readers who don’t think quantitatively.)

One of those numbers that I’ve always been intrigued with is how much of our nation’s total energy consumption relates to buildings. That sounds simple enough.

We are fortunate in the United States that the Department of Energy (DOE) tracks all sorts of energy statistics through the Energy Information Administration (EIA), so a numbers geek can go hog-wild digging as deep as he might want into whatever aspects of energy production and consumption are of interest.

On the energy consumption side, the most common breakdown of end uses is Residential, Commercial, Industrial, and Transportation. The EIA shows the 2011 breakdown of these end-uses as follows: residential 22%; commercial 19%; industrial 31%; and transportation 28%.

In units of energy, total U.S. primary energy consumption in 2011 was 97.3 quads (one quad equals one quadrillion Btus — that’s 1 followed by 15 zeros). Of that, the residential sector totaled 21.6 quads, commercial 18.0 quads, industrial 30.6 quads, and transportation 27.1 quads.

Site energy vs. primary energy

These fractions and quads are of primary energy use. When we use a kilowatt-hour (kWh) of electricity for lighting in our homes, that’s a kWh at our house (site energy), but to produce that kWh of electricity probably took about three kWh of primary energy. The difference between site energy and primary energy is due to waste during the production and delivery of energy to our end-uses. With electricity that difference is huge, due to waste heat produced in power generation and losses in transmission. With other energy sources, such as petroleum or natural gas, the difference between primary and site energy is much less.

So, when we look at this end-use split (22%, 19%, 31% and 28%) we’re looking at the primary energy, and 41% of that (22% + 19%), as commonly argued, has to do with buildings. So far, so good.

Having 41% of U.S. energy consumption attributed to buildings is huge. A very similar metric has to do with greenhouse gas emissions. The percentage is slightly different — because some energy sources are more carbon-intensive than others — but the difference is minor. Buildings account for 40% of total greenhouse gas emissions in the U.S. (usually reported as carbon dioxide or carbon equivalents).

These numbers tell us that we need to be paying a lot of attention to buildings if we want to make headway in curbing our contributions to climate change.

Building-related energy consumption is actually a lot greater

But, I contend that buildings and development patterns actually account for significantly more energy consumption and global warming impact than these numbers suggest.

With the other two segments of energy consumption in the U.S. — industry and transportation — buildings and our development patterns also have a big impact.

Some of the industrial sector energy consumption has to do with heating, cooling, and illuminating factories. Digging deeper into the EIA energy consumption data for 2011, we find that heating, ventilation, and air conditioning (HVAC) accounts for 0.7 quads, and facility lighting 0.2 quads. Those energy uses, I suggest, really fall into the area of building energy consumption (as opposed to the process of making steel or glass, for example).

There is also a portion of industrial energy consumption that has to do with construction materials — the energy it takes to make the stuff we use in buildings houses and commercial buildings. This is often referred to as “embodied energy.” A truly thorough examination of energy use by the building sector would include that portion of industrial energy use for making concrete, window glass, steel I-beams, plywood, and the other building products. I would guess that that is easily a quad or two, though I have not attempted to include that information here.

Transportation energy use related to the built environment

With the transportation sector, the impact of buildings — and where we put them — is a lot more significant. Development patterns have a huge impact on the energy expended for transportation. From the 423-page Transportation Energy Data Book – Edition 31, published by Oak Ridge National Laboratory in July 2012, we find that cars and light trucks accounted for 58.7% of transportation energy consumption in 2010, while medium and heavy-duty trucks accounted for 22.3%, buses 0.7%, air travel 7.8%, ships 5.0%, rail 2.1%, and pipelines 3.4% (Table 2.6).

We further find that for household travel, in 2009, commuting to work accounted for 28.7% of vehicle miles traveled, shopping 15.5%, other family/personal business 15.7%, vacation 2.3%, visiting friends and relatives 9.4%, and other social/recreational 13.5% (Table 8.9).

In Table 8.17 we find that in low-density areas (less than 1,000 residents per square mile) significantly more miles are driven per vehicle per day: 31.6 miles, vs. 18.5 miles when the housing density increases to 10,000 to 25,000 residents per square mile and 14.8 miles with densities above 25,000 per square mile.

The bottom line

What I read into these numbers is that where we build our homes and where companies build their offices and factories has a huge impact on the amount of energy we use in transportation. (I wrote about this in a September 2007 article in Environmental Building News, “Driving to Green Buildings,” suggesting that we consider a new metric of “transportation energy intensity.”)

I would venture to say that a third of all of our transportation energy use relates to building location. We shouldn’t ignore this energy when we’re focusing on the energy use of buildings.

A third of our transportation energy use amounts to 9.0 quads. If we lump that into the expanded column of buildings and the built environment in our initial total of 21.6 quads for residential buildings and 18.0 quads for commercial buildings, and include the 0.9 quads of industrial energy use having to do with heating and illuminating factories, that would mean that buildings and the built environment account for 49.5 quads of primary energy consumption, or 51% or total energy use.

In other words, the direct operation of buildings for heating, cooling, lighting, and appliances, along with a reasonable share of energy use associated with getting to and from those buildings, accounts for over half of our energy use — significantly more than the 41% figure that’s commonly quoted.

This means that those of us involved in influencing how buildings are designed and built and where we build — and that includes building owners — share even more responsibility for the nation’s energy consumption (and greenhouse gas emissions) than is commonly suggested. We have a lot of work to do!

I’d love your thoughts on whether this argument makes sense.

Alex is founder of BuildingGreen, Inc. and executive editor of Environmental Building News. In 2012 he founded the Resilient Design Institute. To keep up with Alex’s latest articles and musings, you can sign up for his Twitter feed.


  1. GBA Editor
    Martin Holladay | | #1

    What if all factories were located downtown?
    Thanks for this blog. You rightly point out that it takes more energy to commute by automobile than public transportation, and that better urban planning might reduce commuting times.

    However, the calculation that attributes a portion of energy used by commuters to the building sector is extremely complicated. I think that you are saying that when employees have to commute 20 miles by automobile to the factory where they work, then the fact that the factory is located at the edge of town instead of the middle of town is an energy penalty that should be attached to (or associated with) the building.

    I understand your logic. But is it really possible to locate all factories and large commercial office buildings in dense urban areas?

    I suppose the answer is yes. But there are penalties to this approach to urban planning:
    Locating factories downtown might increase urban air pollution.
    Locating all factories and large commercial office buildings downtown would require the construction of a more elaborate infrastructure of public transportation. Granted, this would provide a long-term benefit to urban dwellers, but it would be expensive.
    Locating all factories and large commercial office buildings downtown in hopes that workers will be able to commute to work by subway or bus will only work if affordable housing exists close to a subway stop or bus stop. Many U.S. cities already lack affordable housing in these locations, which is why so many people live a long ways away from where they work. Moving factories and large commercial office buildings downtown is not enough if the plan does not include a comprehensive overhaul of America's approach to providing affordable housing. If this last point is ignored, implementing your scheme could have the paradoxical effect of increasing the amount of energy used for commuting rather than decreasing it.

    All of these factors are interrelated and codependent, so your transportation energy use algorithm becomes very complicated.

  2. user-1119494 | | #2

    can we change scattered cities brought to us by cheap fuel?
    Very useful insights, I think. And our cheap fuel and transportation has lead to scattered patterns of living and working served by inefficient roads rather than far more compact modes that work well with walking, bicycles, buses, and trains.

    As Martin points out, there are drawbacks to other modes, but I disagree on some parts: heavy industry has become cleaner and cleaner, making it suitable for closer integration with cities; these stereotypical industries are a tiny part of our employment these days; and much of our new industry hardy belches smoke (I drive by a sparkling Agilent park often), but is sited out where no one lives.

    So sure, why not encourage bringing these attractive industries closer so employees can bike or walk?

  3. JustHousing | | #3

    yes, the argument makes sense
    Thanks for the blog post, Alex. It is good to keep/reintroduce this line of thinking. I say "reintroduce" because your post reminds me of the article that set me on the path to doing things very differently, written in 2003 by Ed Mazria. It was called "It's the Architecture, Stupid," and I think it is a must-read for anyone serious about improving the way we design and build structures. Here's a link:
    I believe it was first published in Solar Today, but I read it in Metropolis magazine.

  4. Expert Member
    Dana Dorsett | | #4

    'Tis true!
    An on-grid New Yorker who walks/bikes/subways to work typically has a lower carbon footprint than a more remote off-grid home where the occupants are commuting by car 2 hours each way.

    US-'mericans hate two things: Density, and sprawl, but density is inherently lower-carb, no way to get around it. When personal transportation becomes more heavily electrified along with a lower-carb grid that may change, but don't hold your breath! Foreclosures in further flung suburban developments exploded when gasoline went well above $4, given that commuting in the F-150 started to compare to the mortgage in monthly cash flow terms. Location matters.

  5. dickrussell | | #5

    We change employers more often.
    Over the last few decades, Americans have tended to have more employers over their working lifetimes. The first job might determine where the employee buys a house, to have a reasonably commute time. But if a change in jobs means traveling an additional ten miles each way, that's usually not sufficient reason to uproot the family and relocate. Each job change may mean an upward ratcheting of commute time, and each new commute distance may not seem "all that much longer." At the end, it took me typically 36 minutes to get to work in the morning, but compared to some I guess I was spoiled. But I was unusual; I had the same employer for over 40 years.

  6. GBA Editor
    Martin Holladay | | #6

    Density is good
    There is no doubt that greater density lowers transportation energy use. I'm noting, however, two things:
    It's impossible to come up with a simple algorithm that ascribes an energy penalty to a building based on its remoteness or the density of its neighborhood. Moving an office building from a less-dense neighborhood to a more-dense neighborhood may decrease energy used for commuting, or increase it, or leave it unchanged.
    People who live on the Upper East side of Manhattan and commute by subway to midtown have it easy. But the maids and the kitchen help who work in midtown Manhattan may have a very long commute indeed, and the same problem exists in San Francisco and many other cities. Without addressing affordable housing, promoting density as a virtue is an incomplete policy recommendation.

    Dick Russell makes an important point. Once you buy an apartment three blocks from your job, what happens when you get laid off? No one is guaranteed lifetime employment anymore, except for Supreme Court justices, tenured professors, and the Pope.

  7. leighadickens | | #7

    region-mobility too
    Not only that, but many people aren't lucky enough to maintain employment in the same city or region their whole lives. We're a lot more willing to be "region-mobile" and go across the country for better work, or work period, than we used to be, especially those of us who live in one of the many small to mid-sized cities and towns in this country that have only limited economic opportunity in each one. That, too, is a consequence of lower density: reduced opportunities (because the opportunities are more spread out) and so more need to transplant yourself quite far to go to wherever the opportunities are.

    Moving around the whole country means even less commitment to the home you're living in now, for energy improvements, "net-zero" goals, "net-zero-plus-transportation-energy-too" goals, etc. Means more travel to visit family and friends left behind and such.

  8. Expert Member
    Dana Dorsett | | #8

    The maids & kitchen help use subways too. (response for Martin)
    Ain't no maids & kitchen help commuting from NJ by car to pay mid-town Mathattan parking rates, seriously! The commuter rail & subway services in the metropolitan NY region are really quite well developed. Even most of the Wall Streeters who choose to live on Long Island choose use public transportation, despite the length of the commute. (The maids & kitchen help working in Manhattan that I've known personally all lived in Brooklyn, and rarely used cars for commuting, but would sometimes drive into Manhattan for socializin'.)

    The housing style in NYC is far lower energy than the single-family suburban model, being both smaller, and having far far lower heating loads due to having only one side of the domicile cube as exterior surface (3 sides for the corner penthouse suites.)

    While there's no dead-simple algorithms or models, the gross math on total amount of gross fuel use & electricity use per resident (OR worker) regionally still comes in well under that of less-dense cities, let alone the low density exurbs.

    Many people will move if the changing job situations make the commute too long. But building where it's many miles from even the nearest grocery store, let alone employer, guarantees it's always a long haul. The calculation on some further flung suburban developments were that if the housing & taxes were cheap, people would come, but those development suffered mightily when transportation fuel prices spiked, with limited or no public transportation options. (I wish my commute were shorter too, but changing houses is cost prohibitive for a number of reasons.)

  9. GBA Editor
    Martin Holladay | | #9

    Response to Dana Dorsett
    I never denied the importance of density and public transportation as necessary features of a less-energy-intensive future. I'm simply pleading for better attention to affordable housing. Without attention to affordable housing, cities don't work well -- even when they work well for the upper classes.

  10. dsmcn | | #10

    Carbon Tax Me
    Shifting our tax structure so that only our use of carbon is taxed would result in individuals and businesses finding the best solution for their own situation, because efficiency would equal significant tax savings.

    We collectively would immediately shop for cars and trucks that were efficient, and demand even better alternatives.

    Everyone would clamor for energy efficient homes, and pv arrays.

    Clearly our current political system does not allow appropriate governance that could make such a change. Within our lifetimes it will become clear that this failure threatens the survival of our grandchildren. Too bad ideologues prevent it from even entering our public discourse now.

  11. watercop | | #11

    Other tradeoffs
    The issue of location is truly fraught. Most who know me in person and online know I'm a staunch advocate for greatly reducing home and building energy consumption. That said, transportation time and energy is something of a sore point with us. My wife and I typically log 60+ miles per day, half of which stems from our somewhat remote location...guilty as charged

    What does that location afford us? Let me count the ways:

    1) Cheaper land - we have almost 1.5 acres with 150 frontage on a deep wide waterway. We get the water view and a whole lot of quiet nature - a hundred trees separate us from the sights and sounds of our neighbors lives

    2) Better schools - the urban county comprising the economic core of our region has simply unacceptable schools. We are graced with three academically gifted children, and careful management of our school district's various alternatives affords us high quality academic experiences for our children without resorting to expensive private schools.

    3) Lower crime rates as well as reduced exposure to the everyday nuisances of high density living. City dwellers choose to accept or least tolerate living alongside those who value unmuffled car engines and teeth-rattling stereos. Not a day goes by without news stories about the latest outrages committed by urban thugs. Space and distance isolates us from such offence and I cherish the solitude.

    I'm seriously considering a Nissan Leaf to reduce our transportation cost and energy footprint, and I might entertain a move a bit closer to the urban core if the stars align, but I don't automatically accept that higher density urban living is the be-all, end-all of a green lifestyle.

  12. Alex Wilson | | #12

    Transportation energy intensity
    In laying out the case for focusing on transportation energy use (2007 Environmental Building News article referenced in initial blog), I have not attempted to make a value judgement. Rather, I have made the case that "transportation energy intensity" should be considered a metric of building energy use--like "energy use intensity" or EUI. For an office building (the type of building I analyzed in that article) here's a quick summary:

    Given the assumptions I used (average U.S. commute distance, average mode of commuting, average fuel economy of our vehicles, and a U.S. General Services Administration estimate of average square footage of building per employee), I calculated energy consumption getting to and from work and normalized that per square foot of building per year. I found that for U.S. office buildings on average (all ages of buildings), the transportation energy use was 30% greater, per square foot, than the building operating energy use. For an office built to the then-relevant energy code (ASHRAE 90.1-2004), commuting energy was 2.3 times as great as building operating energy use.

    The analysis was admittedly pretty rough. Though I spent a couple weeks piling over all sorts of statistics, it was still a fairly rudimentary analysis. One of its big failings, for example, is that it only looked at site energy, not primary energy; the latter would be much better, and I'd still like to do that. But the findings surprised me--that the transportation energy is so high. In other types of buildings, like Wal-Mart stores, it would likely be far higher.

    In other words, I argued, this is something that we should be paying attention to.

  13. bencarsan | | #13

    Architecture 2030
    As Ed Mazria said somewhere, our energy problem is an architecture problem. He and the people at have been banging the drums about this for a while. That site has lots of excellent graphs and charts to make the point that buildings are the critical first step in addressing the energy/climate crisis.

    In their analysis the "building sector" is responsible for using 48.7% of the energy produced in the US. This includes embodied energy in construction.

    As they point out, it is important to consider not only how much energy is used but what kind as well. Since buildings use 75.7 percent of the electricity produced, and since the grid relies on coal power plants and has such a low site-source efficiency (typically in the mid 30s I think) the benefits of efficient buildings are multiplied in terms of source energy saved.

    They also make the argument that, yes, we will do tremendous climate damage with oil and gas, but it's the sheer quantity of coal reserves that pose the greatest threat. And since the coal is mostly used to make electricity, cutting back on electricity consumption is critical.

    Recent climate models show soot and methane emissions over the near term are much greater climate threats than CO2. And I think this again points to opportunity in buildings: even people who are using renewables like wood/biomass for heating should be doing everything possible to conserve.

    In the face of such obvious opportunity, it's surprising that Obama's "all-of-the-above" approach to energy doesn't even consider the single most obvious step we could be taking to fix all this.

    Some of the nations top scientists and engineers recently collaborated on a feasibility analysis for what it would it would take to make New York State fossil-fuel free by 2030. It's available at Achieving demand reduction through building efficiency figures prominently of course, and it turns out the whole thing could be paid for through avoided costs.

Log in or create an account to post a comment.



Recent Questions and Replies

  • |
  • |
  • |
  • |