Arizona DOT Study: Compact, Mixed-Use Development Leads to Less Traffic

Image: Arizona Department of Transportation

Does walkable development really lead to worse traffic congestion? Opponents of urbanism often say so, citing impending traffic disaster to rally people against, say, a new mixed-use project proposed in their backyards. But new research provides some excellent evidence to counter those claims.

A recent study by the Arizona Department of Transportation [PDF] found that neighborhoods where houses are closer together actually have freer-flowing traffic.

Researchers compared some of greater Phoenix’s denser neighborhoods — South Scottsdale, Tempe, and East Phoenix — with a few of its more sprawling ones — Glendale, Gilbert, and North Scottsdale. Some interesting patterns emerged.

In the more compact neighborhoods, the average household owned 1.55 cars, compared to 1.92 in more suburban areas. Residents of higher-density neighborhoods also traveled shorter distances both to get to work and to run errands, the study found.

The average work trip was a little longer than seven miles for higher-density neighborhoods; in the more suburban neighborhoods, it was almost 11 miles. Residents of the three compact neighborhoods traveled just less than three miles to shop, while residents of sprawling locations traveled an average of more than four miles. All of this led the more urban dwellers to travel an average of nearly five fewer miles per day than their suburban counterparts.

The density divide also played an important role in transit use. Rates varied from as high as eight percent transit ridership in high-density neighborhoods to as low as one percent in the more sprawling areas.

All of this translated into a reduced strain on roadways in the places that had more people — running counter to one of the strongest objections to mixed-use development. Comparing one suburban corridor to two of the streets in the more dense neighborhoods, the study found that on the more urban streets, traffic congestion was “much lower,” or about half as high (measured by the ratio of the capacity of the roadway to the actual volume of cars on it).

How did more compact neighborhoods manage to have less congestion? It’s not just because residents there drive less overall. Two design characteristics also ease traffic, according to AZ DOT. Fine-grained street networks distributed traffic evenly across the higher-density neighborhoods, while every driver in the suburban neighborhoods was funneled onto the same big arterials. At the same time, improved pedestrian conditions in commercial centers made it easier for some drivers to park once and walk from destination to destination, taking cars off the road precisely in the areas that attract the most people.

The results of the Arizona study may not apply everywhere, due to the state’s extremely spread out pattern of development. The higher-density neighborhoods still only had between six and seven households per acre, compared with between three and four in the lower-density places. As the report notes, “By Eastern U.S. standards, all of these densities are effectively suburban in character.”

But the report controls for a host of factors, strengthening the conclusion that the different travel behaviors were really the result of design, rather than income, say, or the student population.

The Arizona Department of Transportation deserves credit — first of all, because this is a fantastic, thorough, well-timed study, but also for pointing out the important policy implications. The agency’s recommendations include a public awareness campaign about the benefits of mixed-use, compact development; better planning and public engagement tools; and providing incentives for smart planning.

The authors noted, for example, that outdated policies sabotage planning efforts that are beneficial for livability, public health, and the environment in the name of maintaining traffic flow. The supreme irony — in light of the study results — is that these policies ultimately fail the congestion test too:

Local planners and planning commissions are still using traditional traffic engineering approaches to assess the impact of development projects. By looking only at traffic congestion levels on adjacent links, ignoring through travel, and failing to account for the efficiencies of mixed-use development on lower vehicle trip rates and VMT, progressive projects are likely to be rejected or unreasonably downsized.

The DOT also concludes that congestion isn’t always a bad thing, that density is the key to successful transit, and that short blocks are critical for building vibrant, mixed-use places.

  • Guest

    The headline is incorrect – traffic/congestion are measured absolutely. The chart shows that density rises more slowly that VMT decreases – a situation which absolutely, unoquivocly, leads to more vehicles on the road. 

  • Here from the abstract: “The analysis showed that the urban corridors had considerably less congestion despite densities that were many times higher than the suburban corridor. The reasons were traced to better mix of uses, particularly retail share, which led to shorter trips, more transit and nonmotorized travel, and fewer vehicle miles of travel (VMT). ”

  • Jason
  • Ben Kintisch

    I generally prefer to live in an area where I can walk around to do my errands instead of drive. Even if it costs a bit more.

  • shane phillips

    Yeah, as much as I’m completely supportive of greater density and the transit, bicycling, and walking benefits that come along with it (among other things), your headline is false. Per capita VMT is not equivalent to traffic congestion, and for this to be a plausible conclusion the VMT would have to decline at a rate greater than the rate of increased residential density. That’s clearly not the case. While it’s great that VMT decreases, and that is a worthy goal in and of itself, having one person drive 17.17 miles per day is going to cause a lot less traffic congestion than 10 people each driving 9.12 miles per day.

  • shane phillips

    That said, you can’t draw much in the way of national conclusions from this data. They’re basing this off of the greater Phoenix area–not exactly a bastion of walkable urbanism–and even in the more compact neighborhoods they’ve still got 1.55 cars per household. I think if you look at the higher density regions of the country where owning a car is less of a necessity (or at least it’s okay to only own one), you might see somewhat different trends. Whereas in Phoenix it sounds like every household needs a car for any number of daily trips, in places like NYC that’s simply not the case, and when you reach a certain level of density I suspect that the rate of decrease in daily VMT picks back up, and you see much, much lower values.

  • @google-6bd4d5085f3de5f0622280ad4b5e50c2:disqus only if both areas have the same amount of roads, which it appears they don’t based on the study’s conclusion below.

  • Anonymous

    It’s an interesting study, but it seems like it has a somewhat limited application, being in Arizona and having a fairly low highest density of 10+ households per acre. It would be interesting to see it repeated in San Francisco, for example, comparing different neighborhoods.

  • Anonymous


  • Anonymous

    Sorry about that, not sure what happened.

    Angie, I experienced the same confusion as shane phillips and the first poster (‘Guest’).  The placement of the VMT chart as the lead in to the article is what caused my confusion, and it wasn’t until I realized that the real issue with the lower density development is the bottlenecking of traffic on arterials vs high density development’s fine grain street pattern that allows traffic to filter through many different locations.  Also, high density development generally has fewer miles of road per capita, so the way you explained it to shane phillips wasn’t exactly perfect either.  The important thing to take away from this study is that there is less traffic congestion in high density development because VMT per capita is lower AND the higher connectivity of the street pattern results in less concentration of vehicles which is pretty much the definition of traffic.

    Thanks for bringing this study to our attention, though!  Unlike some of the other posters (who seem to think urbanism is limited to only the large cities), I think this could be incredibly useful in limiting the most sprawling new developments and would have the greatest overall benefit to society since it is so much easier to double, triple, or quadruple densities when the starting point is so low.  Very timely stuff indeed.

  • Desert Rat

    If the intent of the study was to illustrate that density does not increase traffic as the title suggests, the corridors selected are not comparable to make any conclusions on these impacts.

    Central – Given that very little people live along the Central Corridor it makes sense that VMT would be lower. People drive to work then go home, there are no extra stops for groceries or shopping. Central runs parallel to I17 and SR51 so it isn’t used as an access street, the light rail eats up all the roads so people use 7th Street/Ave as an alternative, weekend traffic is minimal (Lower VMT has nothing to do with density.) Central is not really mixed use, no shopping, limited residential.

    Bell – Anyone familiar with Bell Road (AKA Hell Road) knows that the reason what VMT is high is because a) people live there b) people shop there c) people go to school there d) people gamble e) it is a major arterial connecting to either SR51 or I17, f) etc.,. There are more opportunities for trips and thereby an increase in VMT (it has nothing to do with density). Bell has more places people want to go not related to work, weekend traffic is high, Bell is more mixed use than Central.

    Tempe is college town most students bike and don’t own a car as with any college towns there is no mix of activities or demographics, very poor comparison.

    The last sentence of this article is perplexing – What would a transportation agency gain by increasing congestion? Seems counter productive to all those Congestion Management Plans and emission reduction programs funded with gas taxes, and what about the safety impacts?

    Who’s version of livability is being considered? People do not move to the Valley to experience urban density many come to escape NYC, Chicago, Detroit, Phili, LA, etc., many enjoy open desert spaces, back yards, less noise and a better quality of life found in less dense areas. Look at the states gaining congressional seats people are moving to sprawled W/SW, the census still shows people moving to suburbs, Atlanta as sprawled as it is still attracts people from Chicago.

  • Mstanger

    Desert Rat, I didn’t know that lots of little people lived along Central–the researchers should have controlled for height.

    Second, the transportation agency doesn’t “gain” by increasing congestion. The point is that neighborhood traffic congestion is a sign of economic and social vitality. Show me a place with no traffic congestion, and I’ll show you a place that’s dead. Furthermore, you will find a lot of research shows that neighborhoods with higher congestion are safer for pedestrians precisely because the cars move more slowly and carefully.

  • Mstanger , feel free to show the research that suggests pedestrians are safer in areas with higher congestion.  I seriously doubt that if they are, it’s for some other reason than a street full of honking motorists and gridlocked intersections. 

  • Karen

    I read another article about the Arizona study on Planetizen earlier today, and I’d like to pass on a link to this readership, as well, about a project in Austria that is addressing this issue of traffic flow and compact neighborhoods: I find the Austrian project interesting, not only because it goes against one’s initial common sense, but because European cities have been densely populated since the Middle Ages. I think there might be something to be learned from the Old Country…


Study: Shorter Blocks May Be the Key to Cutting Traffic in Small Cities

It’s well-established that density and mixed-use development reduce driving. Right? But strategies like those don’t work the same way everywhere, according to new research published in the Journal of Transport and Land Use. While in major cities, denser development is linked to lower rates of driving, researchers found that in smaller cities it might not […]

Taxes Too High? Try Building Walkable, Mixed-Use Development

Smart growth could increase Fresno’s tax revenue by 45 percent per acre. In Champaign, Illinois, it could save 23 percent per year on city services. Study after study has demonstrated: Walkable, mixed-use development is a much better deal for municipalities than car-oriented suburban development. Smart Growth America recently conducted an analysis of research examining the impact […]

Reminder: Just Laying Track Is No Guarantee Riders Will Come

Laying track isn’t enough to build a successful transit system — as some cities are learning the hard way. A slate of new rail projects — mostly mixed-traffic streetcars, but that’s not the only way to mess up — are attracting embarrassingly few passengers. Some of these projects may be salvageable to some extent, but for now, they don’t […]

How Many Trips Are ‘Captured’ By More Diverse Urban Land Use?

Current methods of predicting the traffic-calming effects of mixed-use development are "woefully lacking" and risk underestimating the transportation benefits of more compact, diverse land use, according to a new report from the Transportation Research Board (TRB). An example of mixed-use development in San Francisco. (Photo: Arch. Record) The TRB report is the first to examine […]