Density Reduces Driving (Even At Pretty High Densities)

Michael Lewyn's picture

A few weeks ago, I read a startling claim in an email discussion group.  Someone wrote that at densities above 20 households per acre (roughly that of a streetcar suburb),* increased density had little effect on driving.  If this was the case, the environmental argument for allowing higher densities would obviously be weaker.

After searching for data relevant to this claim, I found a 1994 study by John Holtzclaw (available here).  Based on data from numerous California cities, Holtzclaw created a model designed to quantify the relationship between density and automobile vehicle miles traveled (VMT).**

Holtzclaw concluded that even in areas with minimal transit service, density affected VMT.  For example, in an area with only two buses per hour, a census tract with 20 households per acre drove about 40 percent less per household than one with two units per acre (15,374 per year as opposed to 27,339).  But VMT did not stop dropping at the 20 households per acre level.  An area with 100 households per acre drove 1/3 fewer miles than the 20-per-acre neighborhood (10,028 VMT per household) and one with 500 households per acre drove 40 percent less than the area witih 100 households per acre (5781).

A similar density/VMT relationship existed in areas with generous public transit.  Holtzclaw calculated that in a place with 100 buses/trains per hour within a quarter mile, a census tract with 2 units per acre produced 20,308 VMT per household per year, and a tract with 20 households per acre produced about 45 percent fewer miles (11,420).  Under these circumstances, an area with 100 households per acre drove about 1/3 fewer miles than the 20-units-per-acre census tract  (7637), and an area with 500 households per acre drove about 40 percent less than the area with 100 per acre (4295).

VMT statistics in actual neighborhoods supported Holtzclaw's model.  For example, in the city of San Francisco, VMT per household averaged 11,256 miles.  By contrast, in the Nob Hill/North Beach area (which had 100 households per acre, as opposed to 48 for the city as a whole) VMT per household was less than half that (5519).  In sum, it appears (at least based on Holtzclaw's model) that even at densities of 100 households per acre or more, higher density means less driving.

Moreover, Holtzclaw's model may slightly understate the influence of density on driving.  Why?  Because higher density, by allowing more people to live near bus or train stops, makes improved transit service possible- which itself seems to reduce driving.***

Why does any of this matter?  Because if the most compact neighborhoods drive less than 20 household-per-acre streetcar suburbs, it follows that additional infill in  already-compact neighborhoods, by increasing density, will reduce VMT.  If this is the case, then additional infill is justified on environmental grounds, even in neighborhoods that are already far more compact than the average American suburb.

*Because there are 640 acres in a square mile, this corresponds to 12,800 households per square mile in a heavily residential area.  San Francisco's Sunset District has roughly this density, as do the more compact parts of Upper Northwest Washington, DC.

**The numbers produced by Holtzclaw's model are at Table 7 of the paper.

***Though perhaps not as much as higher density.  For example, in the super-dense neighborhood with 500 units per acre, the difference between 2 buses per hour and 100 per hour produced a change from 5781 VMT/household/year to 4295- a change much smaller than the difference between either neighborhood and the lower-density areas.

Michael Lewyn is an assistant professor at Touro Law Center in Long Island.



Density and VMT

Based on the figures I have seen, higher densities have little effect in reducing VMT after you reach the density of European cities (about 80 people per gross acre). Higher density still has some effect, but there are sharply diminishing returns.

I think Holtzclaw's numbers understate the effect because they only look at VMT of people who live in the denser area. Actually, higher density also reduces VMT of people in surrounding areas.

Eg, the high density of Manhattan reduces VMT of people all over the NY metropolitan area. If Manhattan were the density of Westchester, the people in Manhattan would drive more. And people all over the metropolitan area would also drive more, because the entire metropolitan area would be more spread out.

Michael Lewyn's picture

gross densities and diminishing returns

1. Holtzclaw seems to be writing about net density, so we can't say much about gross densities from his model.
2. Having said that, 80 per gross acre (or 50,000 households per square mile) is the density of much of Queens (Jackson Heights, Astoria etc) and much of Manhattan has twice that level of density (according to charts at the New York County page). Certainly people drive significantly less in Manhattan than in Queens (though of course its partially because Manhattan jobs are close by- so it is hard to disentagle the "distance to downtown" effect from the "density" effect). It seems to me intuitively obvious that people all over the region would drive more if Manhattan's density was limited to 80 per gross acre, but I'm not sure if anyone has studied the issue.

Densities of Metropolitan Areas

I am thinking of densities of entire metropolitan area, not of individual neighborhoods.

Go to and scroll down to see the chart of Urban Density vs. Private Car Travel in 58 Higher Income Cities. There seem to be diminishing returns around 100 to 150 people per hectare. This is the graph by Newman and Kenworthy that I was thinking of when I wrote my comment.

Michael Lewyn's picture

bar graphs don't prove much

I wouldn't rely on a bar graph for this sort of statement. A bar graph can show you a trend, but it can't tell you how strong the trend is. So for example, suppose that at density X VMT per household declines from 4000 to 1000- a 75 percent drop. If you structure the bar graph to focus on much higher levels of VMT (as the Atlantic link does) that drop might seem insignificant. On the other hand, if you draw the bar graph so that there seems to be a lot of space between 4000 VMT and zero, that drop might seem very significant indeed. In other words, to really know how much the difference between Munich density and Hong Kong density matters, you'd have to see something more precise than a bar graph- instead, you'd want the sort of table that Holtzclaw provides, listing individual densities and VMT levels for each metro area.

Not A Bar Graph and Not Distorted

You must be looking at another graph. The one I am talking about is a scatter graph, not a bar graph. It includes all values from 0 to 30,000 VMT per capita with a linear scale, so it doesn't have the sort of distortions you mention.

You can easily estimate the individual densities and VMT levels metro areas by looking at the graph:
About 4000 vmt per capita at just under 100 people per hectare.
About 3000 vmt per capita at 200 people per hectare.
About 3000 vmt per capita at 235 people per hectare.

The graph in the link provided by cigrainger has much more data, and I recommend that you look at that and estimate the actual densities and vmts for different metro areas. It is very clear that there is no correlation at all between density and vmt above 150 people per hectare.

Incidentally I believe that Holtzclaw provides densities and vmt for neighborhoods. I haven't seen studies by Holtzclaw that compare entire metropolitan areas, but maybe there are some I don't know about.

Fuel demand/VMT and urban density

I'm doing research on this very topic! Here's one of the most up-to-date studies on the elasticity of fuel demand with respect to urban density:

Re: Fuel Demand and Density

Thanks for the link. Your figure 1 also shows sharply diminishing returns after about 100 to 150 people per hectare - which equals about 40 to 60 people per acre.

By the time you get to those densities, Private Transport Energy Use has gone down about as far as it can go, so there is not much benefit to higher densities.

Incidentally, you can get these densities with traditional urban design, without high-rises. To use my favorite example, Coeur de Ville in Le Plessis-Robinson is just over 67 people per acre.

I am sure Le Plessis-Robinson has more auto use than this, because it is surrounded by lower density areas. But if you could build a whole city at this traditional European density, you would get VMT down about as far as it can go.

Density of Streetcar Suburbs

"Someone wrote that at densities above 20 households per acre (roughly that of a streetcar suburb),"

That sounds like a very high estimate to me. Streetcar suburbs typically have houses on one-tenth acre lots, which equals 10 households per net acre and less per gross acre.

Assuming 3 people per household, 20 households per acre is 60 people per acre, about the same as the gross density of Le Plessis-Robinson - which is designed as a traditional European neighborhood, much denser than a streetcar suburb.

As I say in my other comments, I think this traditional European density does give pretty much all the VMT reduction that you can get from higher density.

Driving and total GHG emissions

The premise that if driving is reduced then the planet is much better off I question.

Seattle GHG responsibility is roughly 25 tonnes per capita. (Citation below --- figure includes embodied emissions in imports and assumes electricity is carbon free.) Personal driving is about 2.5 tonnes of the 25 tonne figure. Eliminate driving and saving is less than that 2.5 tonnes figure --- people will use more transit requiring new transit vehicles and energy to propel those vehicles and instead of spending money on cars and gasoline will have more money to spend on non-driving GHG emitting consumption, e.g. flying to Hawaii for a winter break.

Much of the thrust behind the 'environmental' push for high density is I think equivalent to an attempt to rearrange the deckchairs on the Titanic. The problem is not driving per se but our high incomes and effective consumption power.

Reference for the Seattle figures: p 37 fig 14

Prepare for the AICP Exam

Join the thousands of students who have utilized the Planetizen AICP* Exam Preparation Class to prepare for the American Planning Association's AICP* exam.
Starting at $245
AICP CTP Storefont Display

The first online AICP* CTP exam prep class

Are you ready to take the AICP* Certified Transportation Planner exam?
Priced at $245 for May exam!
Women's t-shirt with map of Los Angeles

City T-Shirts for the ladies!

Women's Supersoft CityFabric© Fashion Fit Tees. Now available in six different cities.
Book cover of Unsprawl

Unsprawl: Remixing Spaces as Places

Explore visionary, controversial and ultimately successful strategies for building people-centered places.
Starting at $12.95