Quick, list the top few things North Americans like to complain about! Politics, the weather, questionable pro sports officiating... did traffic make your list? It made mine.

People like to blame traffic on one simple, but logical, cause: there are "too many cars" on the road. Opponents of new development, in particular, cite traffic more often than any other issue as a reason for their opposition. And in most places you’ll find a widespread consensus that traffic on residential streets is particularly objectionable. It introduces noise and pollution, and most importantly, it poses a safety hazard. Keep through traffic to major thoroughfares and off side streets, goes the logic. Development approvals, especially for retail businesses, often even come with stipulations about closing access points to ensure that neighborhood streets aren’t affected by those coming and going.

And we’ve gotten quite good at meeting those stipulations. Visit almost any American city or suburb at the peak moment of weekday rush hour, and you’ll find that most neighborhood streets in that city are so empty you could safely lie down in the middle of them and take a decent nap. Congestion is a problem confined to a handful of choke points: busy, limited-access highways and arterial roads. Most of the pavement we've built and committed to maintain sees only sparse use.

The motivation is understandable. And yet, this approach—keeping the cars on a handful of major routes and out of our neighborhoods—carries with it some troublesome trade-offs. In fact, it’s one crucial reason that traffic (and griping about traffic) is such a fixture of life in the modern American city.

That reason is simple: a hierarchical street network, in which most vehicle trips are funneled onto a small number of major routes, is a perfect recipe for congestion. A well-connected grid, on the other hand, is as good an antidote to congestion as will ever exist.

Want to solve traffic problems, make better use of existing infrastructure (and thus save your city serious money in maintenance costs), and protect neighborhood quality of life? Then we’re likely going to have to be open to seeing a few more cars on our neighborhood streets. The good news: the negative impacts of this have been greatly exaggerated—if (big “if”) we get some key things right about street design.

Why City Streets Should Work More Like a Wetland and Less Like a Shipping Lane

Image: Wikimedia Commons

In 1965, a fellow named Mr. Go was born. Or rather, MRGO, the Mississippi River Gulf Outlet, which was dredged by the Army Corps of Engineers to provide ship traffic from the port of New Orleans a shortcut through the river’s labyrinthine, muddy delta and into the Gulf of Mexico.

The drawback to MRGO became all too clear when storms dropped colossal amounts of rain on the Gulf Coast region in a short time. The original marshes south of New Orleans were capable of absorbing the resulting huge influx of water, distributing it across hundreds if not thousands of tiny channels, and slowly but steadily filtering it through. These marshes were New Orleans’s best protection against cataclysmic flooding.

The newly deep, dredged channel of MRGO, on the other hand, would, in flood conditions, become a fire hose aimed squarely at New Orleans. It was a significant factor in the destruction wrought by Hurricane Katrina in 2005, amplifying its storm surge dramatically. In 2009, the channel was closed for good.

A freeway or stroad in an urban setting is like Mr. Go. A complete grid of smaller surface streets, in which there are many equally attractive routes from A to B for drivers to choose from, is more like a wetland, and actually does a great job of dispersing traffic and preventing congestion. Consider the trip from Point A to Point B in each of the below maps:


Suppose that at rush hour, 3,000 drivers in a one-hour period want to get from A to B. In the hierarchical network, they have only two routes available to them. If both are equally convenient, it’s reasonable to think about 1,500 will go each way. The straight, through routes outside of the enclave-like subdivision will need to be built to accommodate that heavy volume of through trips. Thus we get the suburban default: “pod” neighborhoods flanked by wide, busy thoroughfares.

In the traditional grid, on the other hand, there are many possible routes using combinations of the five north-south streets and five east-west streets. (To be precise, there are a total of 70 different routes from A to B.) Assuming these routes are equally attractive, the traffic will “soak in,” dispersing among the various routes. Now, no one street carries a high enough volume of cars to create the imposing moat-like effect of a suburban arterial road.

Even better: the grid has built-in resilience where the hierarchical layout has built-in fragility. Imagine there's a crash that temporarily closes the street segments in red below:


The total number of possible routes from A to B drops from 70 to 54, but traffic isn’t terribly adversely affected. Drivers adjust by taking detours down different side streets.

In the hierarchical scenario where all traffic is funneled onto one of a couple stroads, on the other hand, a crash on either of those stroads brings everything to a grinding halt.

What happens in a hybrid scenario: a complete, gridded street network, but one in which the major thoroughfares are designed with wide lanes and a high speed limit? Let’s say in the below diagram, a modified version of Scenario 2, the streets drawn in bold can accommodate traffic at 45 miles per hour, while the interior streets are limited to 25 miles per hour:


Traditional grid with higher-speed thoroughfares in bold: Scenario 3.


Drivers trying to get from A to B will find it faster to use the arterial routes, and so they will… until, economic theory says, rush-hour congestion hits a point where those arterials are only flowing at an average of 25 miles per hour. Once they offer no speed advantage over the side streets, drivers begin “rat-running”—cutting through the neighborhood to avoid congestion on the main drag.

The conventional response to this rat-running problem is too often to close off neighborhood streets to through traffic, by the use of bollards or forced turns or any number of other barriers. Then, we widen the through roads to make sure they can accommodate all the traffic that has been forced onto them.

The correct response is quite different, and counterintuitive. We should actually consider slowing down traffic on the main thoroughfare. Narrow the lanes, add street trees, bumpouts, and other traffic calming devices. Maybe even remove lanes.

Why? Because if one route from A to B is significantly faster, in free-flow conditions, than all the other routes, it will predictably become congested at times of heavy traffic. If all of the routes, on the other hand, are roughly equally attractive, then traffic will disperse itself and no one street will suffer an inordinate amount of it.

It Works in Real Life: A Look at San Francisco

It’s always good to be skeptical of argument-by-economic-theory. Reality is messier. But there are real-world examples we can study of the power of distributed grids. One such example is San Francisco, a city which:

  • has a very complete, well-connected grid of streets.

  • doesn’t feature strong distinctions between “through streets” and “neighborhood” streets.

  • most important for our purposes, makes traffic-count data readily available to the public.

Here is a map of a portion of San Francisco’s Pacific Heights neighborhood bounded by Divisadero, Green, Laguna, and Sacramento streets. Within this rectangle, each block of each north-south street is labeled with the average number of cars per hour measured on that block during the afternoon peak period: 3 p.m. to 6 p.m.


Click to view larger. Data source: SFMTA.


Part of the featured section of Divisadero (Image: Google)

You’ll notice that the closest thing to a major street here is Divisadero Street, the leftmost edge of the rectangle, whose traffic counts range from 736 to 961 vehicles per hour—but even so, the volume of traffic on Divisadero is not that much greater than on the side streets. It’s about double. Otherwise, traffic is spread remarkably evenly throughout the neighborhood.

What if we were to take this same number of vehicles and send them all up and down Divisadero, by doing something to block off the other north-south streets? This is to say, what if we applied a typical suburban traffic pattern to San Francisco, in which there is an ironclad distinction between through streets—spaced half a mile or a mile (or more) apart from each other—and neighborhood streets?

The traffic counts in that scenario would look like this:


Click to view larger.


Accommodating over 3,000 vehicles per hour requires a four-lane (or wider) street. Divisadero would become a moat of cars dissuading residents to the east of it from walking to businesses, parks, or friends’ homes to the west of it, and vice versa. This would further inflate traffic levels by driving more people into cars who might otherwise have walked, in a vicious cycle. Rush hour congestion would become a bigger issue. More land would have to be devoted to parking instead of productive uses. And all of this automobile orientation would put a damper on the economic productivity of the neighborhoods flanking the street.

Tackling the Trade-Off

In the traditional urban approach, exemplified by San Francisco, there isn’t a rigid distinction between through streets and neighborhood streets. The trade-off of this: you won’t find any streets with virtually zero traffic, as so many suburban homeowners have come to expect.

The notion of dispersing traffic throughout the grid in a place that, unlike SF, hasn’t been operating in that way thus tends to violate expectations and to meet with resistance. Do homeowners need to sacrifice their quality of life by allowing through traffic to disturb the peace of their neighborhoods, just to avoid congestion on a few busy roads? In a word, no.

Here are a couple ideas to move the conversation past these initial fears and on to constructive alternatives:

Break Down the Numbers. Often, numbers related to the traffic impacts of development are big and scary when not put in an intuitive context. As an example, a development proposal for a condominium complex with 180 condos adjacent to an area of single-family homes has been met with fierce resistance from those homeowners over traffic impacts. The objecting neighbors want a roadblock put up to keep condo residents from cutting through the neighborhood. Their refrain: “The traffic study says that’s over 750 new trips a day! It’ll ruin our quiet streets!”

Let’s reframe. 750 trips a day, in this case, translates (according to trip-generation tables—which are pretty unscientific, but that’s a topic for another article) to an estimated 68 trips during the busiest hour in the afternoon. That’s slightly more than one vehicle per minute. And less than that every other hour of the day. Perhaps that sounds less scary than 700?

In the San Francisco example above, the afternoon rush-hour traffic on the north-south streets examined ranges from an average of one car every 17 seconds to one car every 4 seconds. That’s not insignificant—but that’s rush hour, in one of the most densely populated cities in North America.

Need even more convincing that traffic is manageable when it’s distributed across a grid? Strong Towns contributor Andrew Price, in an article he wrote on traffic and development, shared this video from Manhattan, the most densely-populated place in North America:


Identify the Real Issue. Is it actually traffic volume that is the sticking point with a proposal that might send more drivers down side streets? Especially if that volume comes out to something like a car per minute? Or is it something else, like speed (and therefore safety), or noise?

If it’s a more tangible issue like that, the best solution may come down to something we have long advocated at Strong Towns: respect the difference between a city street and a high-speed road. Streets are platforms for human activity, and we need to calm traffic on them until it can comfortably coexist with humans on foot.

No one should be cutting through a residential neighborhood at 35 miles per hour. But at, say, 20 miles per hour, it’s possible to have a distributed grid of safe, civilized, productive streets. Such a grid can handle traffic in a resilient and cost-effective way. And it can help free us from the awful tyranny of the overbuilt stroad.

(Cover photo: Wikimedia Commons)