« Countdown to Impact | Main | Bullseye »

Density as Efficiency

EnergySavingsNUN.jpgNew Urban News has an all-too-brief article about research by John Holtzclaw of the Sierra Club and Jennifer Henry of the US Green Building Council comparing the energy efficiency of "high-density urbanism" to Energy Star-rated homes. The result was surprising, even to people already inclined towards dense urban environments: even the maximum Energy Star savings was beaten by moderately-dense development of 12 housing units per acre. At 48 units per acre -- a moderate apartment or condominium complex -- the energy savings were double that of maximum Energy Star. The savings arise largely from efficiencies in infrastructure and transportation. The combined effect of higher-density living and usable non-auto transit is called "location efficiency."

Some related findings:

• An average urban household uses 320 million British thermal units (mBTUs) annually, while an average suburban household uses 440 mBTUs (assuming 2.5 people/family). The difference is mostly in transportation and infrastructure.

• Access to transit yields significant energy savings, but not as large as increased density.

• The economic savings from enhanced location efficiency from 10 years of new construction are about $2.3 trillion, mostly from reduced auto ownership, according to a study by Holtzclaw with David Goldstein and Mary Jean Burer of the Natural Resources Defense Council.

A number of questions arise from this line of research.

Energy Star, even at its maximum deployment, is not as radical as some other high-efficiency building and appliance systems. How much could household energy savings be improved using design elements such as R-2000, phase-changing wax insulation, and White LED lighting?

How can communities now heavily dependent upon autos transition to more energy-efficient characteristics? Another article in the same New Urban News issue looks at some encouraging changes in the Los Angeles region, and cites the city of West Hollywood as an example of what can be done. (I lived in West Hollywood for a few years, and I can attest to how enjoyable it was as a walkable city in the heart of LA.) What steps have the best payoff in terms of encouraging location efficiency?

Single-family home ownership remains (at least in the US) as a goal for many people. How can higher-density urbanism attract that same desirability? What aspects of higher-density urbanism are in most need of re-evaluation?

As we think about ways to make higher-density urbanism more attractive, how can we reduce its financial cost? Many of the American cities already home to higher-density residential urbanism are also among the most expensive. To what degree is that a function of too little supply and too much demand, and therefore mitigated by increased higher-density urbanism in periphery locations?

How much of an improvement would come from applying Energy Star (or better) efficiencies to higher-density urbanism? That is -- just how good could we get, if we really tried?


Listed below are links to weblogs that reference Density as Efficiency:

» Speaking of density from Gil Friend
WorldChanging.com published a piece the next day, coincidentally, on Density as Efficiency , reporting on research comparing the energy efficiency of "high-density urbanism" to Energy Star-rated homes. [Read More]

» Packed Like Sardines – Density is Good from Treehugger
WorldChanging has a good post about this New Urban News article. It is about the energy savings that come from urban density. Quite interesting. For example, even homes with maximum Energy Star savings get beaten by the relatively sparse density... [Read More]

Comments (20)

I like New Urbanism (and of course energy conservation technologies) very much -- It seems that as usual the biggest barriers are politcal and cultural.

Just using off the shelf designs would be a tremendous step.

It would be great. So how to convince politicians, developers, planners, who as a rule view multi-story parking lots as a cypher for progress?

That is a question I don't have an easy answer for. Raised awareness helps.

It's great to see some stats for what I think is the most powerful argument for denser living (and maybe the only one that can gain a wider consensus) - that it's not about 'friendly neighbourhoods' a la New Urbanism (a movement that I have to say makes me slightly nauseous) but that it's all about energy saving and economic sustainability.

Your point about how to make denser living more attractive is exactly right - because we have the right to be able to live a lifestyle that is actually appealing and not a forced choice. I would say (as an urban designer) that it is all about design - cleverer use of space, more innovation, more local distinctiveness, and certainly not the off the shelf designs posited by the previous comment.

Hana - it is fine to appeal to your own authority, but I am left unclear WHY you dislike New Urbanism in particular (there are other "high density" motifs of course) versus your own designs.

I'm actually fine with any living arrangement that raises density and conserves energy.

The whole point of using something off the shelf is it has a better chance of "sticking" - 15 odd years of work and actual communities is a good starting point, versus maybe doing it wrong.

Off the shelf doesn't mean you ignore local distinctions and artistry.

I'm not talking about building a better Burger King.

I wonder if "density" is a proxy for something else. Certainly there are people living in rural settings consuming little energy and resources - for instance in China, India, Bolivia, Mauritania, and so on. Certainly there are people living at high densities, inside upscale gated communities for instance, using tremendous amounts of resources. People live at high densities in Houston and Stockholm, yet people in the latter use far less than the former.

So we should study dense settlements closely, and discover just what it is about them that makes efficient living easier. Iit's not density per se that matters, but having jobs and shopping close by, and having enough people in an area to provide efficient mass transit. Policies should encourage these advantages, not just dense settlement patterns.

There are also examples of people living at high densities in rural areas. For examples, visit the thousands of agricultural villages still dotting France, Spain, Denmark, Italy and other countries.

So many great questions, great comments, and a great topic. I'm still mulling it all over.

Quick note, though, to David - I'm not sure if you have subset data for Houston and Stockholm which is different than this, but per capita energy consumption in the United States and Sweden are roughly the same.


I meant to say *residential* energy consumption, David.

I think (based mainly on living in New York) the two biggest things attracting people towards owning a home and away from living in dense urban areas is:

1) a belief that it's safer and "better" to raise a family in something like suburbia, rather than in a city. Sanitation, cleanliness, and paying some economic attention to slum/ghetto areas could help in this area, plus playing up the positives for children in cities (they're really great places to raise families, think of all the culture and friends and things)

2) the desire to own one's living space and be able to make modifications and improvements to it. Dense urban spaces mostly offer only renting, or very expensive owning, and sometimes even if you want to make an improvement to just your part of a building, you need to get approval from a co-op committee or building overseer. Not sure what can solve that. Another aspect of this is the density itself - efficiency is great, but how much do we want to mechanize ourselves in the name of efficiency? I mean, sometimes city living engenders cabin fever, and having parks nearby doesn't always help.

Thanks for your comment Joseph. I think the data you cite are wrong. Swedish per-household residential energy use in 1999 was .898 metric tons of oil equivalent (TOE) or 38,193,151 Btu. In the Southern U.S. (which would include Houston), residential energy consumption per household in 2001 was 2.1 TOE, or 89,500,000 Btu. (Sources, US DOE Energy Information Agency, CIA World Yearbook, Earth Resources Institute.) Swedish homes are built to a better energy standard than the Canadian R2000 program, which is itself pretty darned good.

PS - Joseph, I hope you realize that I mean no disrespect by the above comment. I studied your reference, and I'm puzzled by the numbers, since many of the European data contradict my understanding from other sources. Here's a link to data from the US Energy Information Agency:


Here is a research paper by John Holtzclaw, showing the relationship between residential density and automobile miles traveled: Smart Growth -- As Seen From the Air. Check out the graph on page 8, which shows you really don't need a huge level of density to gain major efficiences in travel. Even neighborhoods with sideyard houses, courtyard houses or townhouses can cut vehicle miles in half, compared to standard suburbs. Here's a list of resources on cities and transportation from the Sierra Club.

The Lincoln Institute's Visualizing Density website is a great resource for understanding what density looks like. It's a database of aerial photos showing U.S. neighborhoods at a wide range of densities. The database is also available as a report by Campoli and MacLean (free registration required).

Here are some of my own thoughts on density and urban design: Delightful Density

Something to keep in mind when studying density is the difference between residential (or net) density, and total (or gross) density. The former is density on residential land, and the latter is density on all land. That makes a big difference, because of all the non-residential land uses like parks, streets, commercial/civic buildings, parking lots, etc. Holtzclaw uses residential density in his research, while the Census Bureau reports gross density. There's no standardized conversion from one to the other. For any given neighborhood, the residential density statistic might be double the gross density statistic -- or more.

David, I'm just not seeing where you're deriving your numbers. The data I referenced is pulled from "Energy Balances of OECD Countries", which unfortunately cannot be accessed directly (unless one wants to throw down a couple hundred bucks).

I found an Australian government site which uses the same source and shows pretty much the same thing:

The only difference is that the data is slightly older and in their data set, Sweden actually consumes slightly more per capita in residential use.

That link you provided to EIA data doesn't seem to have residential sector data. And I am not having luck dissecting Swedish energy statistics (since they're mostly in Swedish).

I don't doubt that the Swedes might be more efficient in use, but their climate would tend to lead to high residential consumption.

As I said earlier, it would be pretty hard to get down to the detail level of Houston v Stockholm. It's safe to assume, though, that since Stockholm's climate is more temperate than Sweden's as a whole, and Houston's is more energy-demanding than the US as a whole, coupled with density differences of Stockholm vis-a-vis Houston, then one would likely see the figures diverge from the rough parity one sees at country-level data.

I did get down to state level with residential energy consumption, and Texas only consumes about 3.7% more residential energy per capita than the US as a whole, so actually the national figures for the US might be roughly representative of Houston.

My apologies for quibbling, but I get curious about assumptions we tend to hold. It would be interesting to get a more definitive answer, since your choice of cities is a good one (relatively dense Nordic versus sprawled in humid southern climate).

Laurence, thank you for those links.

David, Joseph -- I want to thank you, as well, for demonstrating how disagreements don't have to lead to insults and hostility in these discussions.

I've been playing around with household vehicle data,


and there's some interesting conclusions from it.

* Over eighty percent of vehicle miles traveled in an average household are for five types of trips: 1) To/from work, 2) Other family/personal business, 3) Shopping, 4) Other social/recreational, and 5) Visit friends/relatives.

Of those five trip types, only the 5th one (and maybe part of the 2nd one) cannot be substituted for very easily (ie, there are too many variables and you can't easily get your friends and relatives to cluster in proximity to one another).

* The average household spends 1,176 person-hours in their vehicles per year, which comes out to 441 hours per person. That translates in to about 1.2 hours per person per day in a household vehicle.

* There are about 180 million household vehicles in the Unites States, or 1.7 vehicles per household. The average vehicle is used for 1.07 hours per day.

Also, I compared our two main cities (Minneapolis and St. Paul) to a typical outer ring suburb (Plymouth).

Minneapolis is about 3.5 times as dense as Plymouth, and St. Paul is about 2.5 times as dense as Plymouth. Densities in this metro seem fairly average for a US metro.

Even at Plymouth's density (keeping in mind that about 10% of the area around here is water), there are 1,951 people per square mile. I often work with a concept I call a "wiwaldi unit" - "wiwaldi" being short for "within walking distance". This unit is 1/8 of a square mile - which is basically a diamond-shaped area in a standard city grid where the middle point is within five minutes walk to the edge of the unit.

Plymouth has 244 people per "wiwaldi unit".

So, the question comes down to - is it possible to substitute for those 80% of vehicle miles traveled by making the purposes of those vehicle trips accessible by foot?

What I've been working on for many years is answering that question -- meaning how to take advantage of the fact that the average vehicle only gets used 1 hour per day, and that probably 90% of vehicle miles traveled can be substituted for by walking-distance solutions.

It's a fairly involved set of solutions, so I won't go in to it here, but needless to say it is easily possible to do with existing densities of an average city like St. Paul, and might be more easily done in suburbs as time goes on (or at least in specific pockets initially).

I don't think it's good to try and appeal to people on the basis of "density" or "urban" living, since both of those words have some extremely negative connotations. What people are trying to do is live well, and doing that involves many things, of which many have already been mentioned upthread -- affordable, controllable, safe, enjoyable, etc.

Living in a given urban area is not unlike being in any public space -- there's lots of negative behaviors and environmental conditions which one cannot control. Moving to a bigger home with more land, and spending more time in a car -- all of that is about controlling one's environment and limiting exposure to negative things. Gated communities only extend that approach by making explicit limitations on behavior and the environment, and making that enforceable, where in a denser urban setting, this usually isn't possible.

It's not about an "urban" or a "dense" solution, but rather a "living well" solution that needs to be clearly superior to any other option. And that option needs to be available anywhere - not just in a core city or a "new urbanist" section of a suburb.

Belatedly replying to the question about New Urbanism - my main issue is the assumptions underlying NU about what constitutes a 'good neighbourhood' and why, and how these 'lessons' can be applied to the many different urban situations that one finds. I'm just not sure that there really exist good 'off-the-shelf' solutions and unfortunately, although I know that there are many sensitive designers working in the field, I do find that a lot of the time NU ends up being about better Burger Kings and not about local distinctiveness and creativity. The problem with design codes is that local authorities see in them the possibility of a ready-made solution and apply them willy-nilly in situations which are not appropriate. And I simply don't see how Seaside, Florida is a good model for any sort of sustainable urban life, given that everyone has to drive to get there, its monocultural white middle-class demographic and so forth.

I don't want to see place-making that is a parody of the past - I want to see places that are of their time and respond to our diverse, provocative and often difficult culture. Despite their protestations, I don't see the New Urbanists really responding to this, instead creating a hollow pastiche of a social culture that I'm not sure we all really want (see http://virtualhana.blogspot.com/2005/06/myths-of-social-capital.html).

I'm not arguing that I may be a better or worse designer. But I do see other places that have created denser, more sustainable communities in the lasts fifteen years that respond more appropriately (eg Borneo Sporenburg in the Netherlands) than a NU development on the suburbs of a city (inherently already making it unsustainable) and built for a narrow social demographic. Yes, I want to see more energy-efficient human settlements - but my personal opinion is that in NU, a nostalgia for archaic social patterns matters more than energy-efficiency - otherwise they would be building much more densely, in inner-city areas, and using materials and construction technology that is way more innovative.

And I completely agree with Joseph's point that you can't appeal to consumers on the grounds of density or urbanity, and that it's living well that counts. I would add, that to policy-makers, the energy-saving argument will be the strong one in terms of getting higher densities enshrined in planning codes (as they are now in London).

Don't know if anyone is aware, but London's example is pretty interesting as they've derived a density code that relates to how far you are from public transport - you get this really crazy map that has combined the walking distance from every underground (metro) and bus stop and gives you a minimum and maximum density accordingly. Pretty neat and makes a lot of sense. It's called the PTAL (Public Transport Accessibility Level) and there's more detail in the London Plan (free download here http://www.london.gov.uk/mayor/strategies/sds/index.jsp)

Seaside is on my links list of traditional neighborhood developments, as is Java Eiland in Amsterdam. I don't list Borneo Sporenburg because apparently it lacks mixed use, essentially being a high-density bedroom community. The developer cut a deal with a local shopping center operator so there's no mixed use on Borneo Sporenburg's streets.

Seaside is a walkable, bikable community where people drive far less than in the usual condo-tower resorts lining the Florida panhandle coast. The lots there originally sold for $15,000 -- an affordable price for most Americans. It has a wide range of uses, from shops to a school, offices and a church.

On my links list there is everything from low density rural hamlets to high density downtown revitalization. What they all have in common is that they're compact, walkable and diverse. More than half of new urban projects are urban infill. New urban developments increasingly employ green building techniques and other sustainabilty strategies.

I don't advocate any return to a 1920s social structure, and neither do new urbanists generally. What's more, urban design doesn't have the power to effect such a change. There are relationships between urban form and the people who live in a place, as well as behavior -- not determinative impacts, but indirect influences and feedback loops. The papers "Reconciling the Link between New Urbanism and Community" and "The Social Goals of New Urbanism," both by Emily Talen, are good discussions of these issues.

Jack Sam:

A bit on Semantics and Simulacras:

It seems that the commenters to this post want to move away from the negative connotations of 'urban' and 'density' to the more desirable idea of 'living well'. I think it is important make all those terms synonymous rather than to ditch those aren't currently positive. Here's my example.

I'm a 22 year old Chinese-Born-Canadian. My teacher and our class got into an argument for using the term 'multi-culturalism' to describe the acceptance and integration of diverse cultures which Canada is by and large known for.

For a man who exposed me to Jean Baudrillard and his notion of Simulations and Simulacras, it surprised me that he could not get over the fact that multi-culturism was a big flop in Piere Trudeau policy that wasted a lot of money in trying to teach both English and French in Canada. Because I'm young and I don't have much of an interest in Canadian history, I didn't know that this was the root of 'multi-culturalism'.

But who cares ? That is not what the term means to me, or the rest of world anymore. My parents, and the parents of MANY of my friends immigrated to Canada because of the contemporary accepted positive connotion.

So my point is, it would be better to holistically sell the notion of urban desnity as living well because, one day, those negative connotions will be forgotten. It would be easier to achieve massive buy-in even from the most progressive thinkers for one reason or another may still harbour negative connotations to words that should mean good things.

Stewart Brand:

Good discussion here, great topic.

I'm particularly interested in the one billion people living in squatter cities worldwide (with two billion more coming).

The squatter cities are exceptionally dense, especially compared to the rural communities the occupants left.

Does anyone know of research comparing the environmental effect of the slum dwellers versus that of the villages they left? This would not be just a matter of fuel but also of firewood cut, bush meat killed, marginal agriculture, etc.

(As it happens I've lived happily for 22 years in the densest housing in California--- a squatter neighborhood of 400 houseboats in Sausalito.)

Stewart, here are some pieces that may help you assemble that puzzle.

Geographers and ecologists are just waking up to the existence of village landscapes as a unique category of land use. These are agriculturally productive areas that are denser than rural lands, but less dense than urban areas. The village landscapes of Asia are home to one-quarter of the world's population, and release more than half of global greenhouse gases from agricultural land and biomass fuel.

Research by Erle Ellis has found certain improvements in China's village landscapes. In "Long-term ecological changes in the densely populated rural landscapes of China," he writes,

"At all sites, many of the less productive crop lands have been abandoned in response to changes in land policy, increased wealth, and as populations leave village landscapes for the city, either temporarily or for good... It therefore appears that, contrary to expectations, land transformations associated with population growth and the adoption of modern technologies are driving substantial improvements in ecosystem services across village landscapes."
In China, Ellis says, "I would guess that the slums, not exactly squatter cities, are a local environmental problem, but are part of the general trend toward abandoning low profitability agriculture on marginal lands, which is leading to perennial vegetation regrowth across
large areas of rural China -- globally, regionally, and locally a good thing for many reasons."

That's an interesting twist, Laurence. Thanks for the info.


This page contains a single entry from the blog posted on July 3, 2005 2:52 PM.

The previous post in this blog was Countdown to Impact.

The next post in this blog is Bullseye.

Many more can be found on the main index page or by looking through the archives.

Powered by
Movable Type 3.34