Are one-way streets a plague on mankind?

Or at least a plague on Louisville?  From Emily Badger in The Washington Post:

…they took advantage of a kind of natural experiment: In 2011, Louisville converted two one-way streets near downtown, each a little more than a mile long, back to two-way traffic. In data that they gathered over the following three years, Gilderbloom and William Riggs found that traffic collisions dropped steeply — by 36 percent on one street and 60 percent on the other — after the conversion, even as the number of cars traveling these roads increased. Crime dropped too, by about a quarter, as crime in the rest of the city was rising. Property values rose, as did business revenue and pedestrian traffic, relative to before the change and to a pair of nearby comparison streets. The city, as a result, now stands to collect higher property tax revenues along these streets, and to spend less sending first-responders to accidents there.

Gilderbloom and Riggs have also done an analysis of the entire city of Louisville, comparing Census tracts with multi-lane one-way streets to those without them. The basic pattern holds city-wide: They found that the risk of a crash is twice as high for people riding through neighborhoods with these one-way streets. The property values in census tracts there were also about half the value of homes in the rest of the city.

The full story is here.

Comments

I can believe this. One way streets are extremely frustrating; miss an address and you have to take a long way around. Drivers are more inclined to do stupid things ( like backing up) to avoid a big go around.

Worse than one-way streets are the limited access highways that cut through cities, dividing cities into separate, and often very unequal, districts. What were they thinking? They weren't. The problem with "experts" in traffic is that their focus is on how to move people from one place to another, not how to promote economic and social development.

There is a problem with "experts" in general. Hayek talked about it. But, we should give experts twenty trillion dollars to combat global warming, because expertise.

Impressive.

From a jab at intellectuals (citing an intellectual, no less) to a jab at progressives in 2.5 sentences!

"The problem with “experts” in traffic is that their focus is on how to move people from one place to another,"
-That's not a problem, rayward.

It is a problem when the act of transporting some people negatively affects others via an externality. (Noise, stress, danger, pollution, congestion) Somehow the two need to be balanced, focussing in transport only suggests that cities only exist to be conduits for cars and that everything else is irrelevant.

"dividing cities into separate, and often very unequal, districts."
-So you don't want the poor and the rich to live in the same cities?

Why would highways increase inequality? Would two unequal neighborhoods separated by a highway be a single homogenous neighborhood if the highway wasn't there? That seems unlikely (and is contradicted by evidence from cities that don't have highways). Also, how significant is the dividing effect? Bridges over highways exist. And many cities are naturally divided by features like rivers. It's not clear why an artificial division would be worse.

It's true that highways in cities tend to be aesthetically displeasing, but their harm beyond that is not obvious. They also greatly increase the convenience of travel for people living in or visiting the city. It's not clear that they are a bad thing overall.

"Would two unequal neighborhoods separated by a highway be a single homogenous neighborhood if the highway wasn’t there? "

Yes, and there were plenty of natural experiments proving this in the 50s and 60s. (I'm sure there are some cases where this was not true as well)

As far as the accident statistics, this seems to be a common phenomena: the more attention and care required of drivers, the fewer accidents and the slower they drive. Narrowing streets, even visually by adding planters that extend locally to the travel lane, has a similar effect. Sizing of most suburban and urban roadways are dictated by fire department requirements for heavy equipment access, along with some bad traditions in the US, and are overly wide and have excessive corner radii compared to what an up-to-date traffic engineer would currently design for normal mixed traffic.

Accidents would be even lower if we made everyone walk

Why not just say that each person needs to pay the actual costs of their transport. The real estate involved for their conveyance (right of way) the improvements on the real estate (paving, running rails etc) and the operation. It would be a free market dream and walking would be the cheapest, easiest and fairest option for all.

One-way streets allow you to do away with the problem of cars blocking the whole street by lining up to turn left.

I am not sure I agree. Two streets could convey a result due to some random chance. Overall for narrow roads, one way ensures faster traffic movement. As for the increase of pedestrian traffic and other stuff it all seems hocus pocus.

You should read more about traffic engineering before making a comment based on your gut. Besides, is "faster traffic" the goal of cities?

Why don't you tell us the goal of transportation systems, if not to transport people (distance over time)?

I think we've learned that goals of transportation systems need to be balanced with other goals of cities. Lots of natural experiments in this the last 60 years.

The purpose of a city is not to transport people. It's to be a place to live, work, etc. Part of that is transportation but if transportation takes precedence over everything else and affects others through externalities then you get some terrible outcomes.

I don't see if they normalized for throughput. You would want to normalize the number of collisions and injuries for the throughput not the 'number of cars on the streets' (whatever that is supposed to mean).

As for property values, my guess is that higher throughput streets have lower property values. I don't see any other simple explanation of the relationship between one-way streets and property values.

Reading through the linked article it sounds like the authors have a religious anti-car agenda. My initial guess is that they know the proper metrics and they will simply omit them for their clickbait paper.

The other question I have from a methodological standpoint is this:

What caused the planning agency to move from single lane to two lane traffic.

You get a different story if, for example, the planning agency saw an increase in traffic before the change took place, or the planning agency changed policies because of planned urban density arising from a new shopping center, or high rise, that was being planned to be in existence in two years, etc.

What caused the change for new traffic patterns, and was that correlated with increased traffic.

That's a great question. Did they do it randomly (seems unlikely), or did they do it in response to some change in the environment?

One note though, it seems unlikely that two way streets would have the throughput of one way streets. Unless, something else changed, like the roads were widened.

Or that there was a new connection someplace else in the network that expanded capacity at a node.

Rather than "anti-car", why not, "anti-unpriced transportation externalities" or "anti-road subsidy"

The challenges whose solutions they are seeking are all related to quantities of cars, not the existence of cars. As long as we refuse to price the real estate on the street, we will get more demand than we know what to do with. Price it correctly, with with actual prices or congestion, and you can start to get something akin to balance.

One thing I have learned from reading transportation debates over the years is that people who complain about a "religious anti-car agenda" always have a religious pro-car agenda.

LOL I didn't even know people like you and Stuart existed. Always learning

It is so simplistic to say it is one way streets. What if it were two two way streets and the rest were one way? Speeding is not dangerous. Speeding can be dangerous when the speed differential between one car and the rest of the traffic is larger than the reaction time etc...

"Speeding is not dangerous" is a rather simplistic view of a more complex phenomena. Traffic averaging 35 in a residential area is more dangerous than traffic that averages 25, all things equal. (That v^2 thing.)

Along with speed, one way streets convey a sense to drivers that they are in a protected environment - and people respond to incentives.

Talk about simplistic. Talk about incentives.

Why don't you talk about how too-low speed limits encourage speeding and therefore the speed differential?

Who said anything about speed limits? I certainly didn't. I know the literature fairly well, btw.

Not really sure what your point is.

what speed are you travelling now?

People do respond to incentives. What is the incentive in your scenario?

When you make roads/driving feel more safe, people respond by taking more risk.

Peltzman effect
http://marginalrevolution.com/marginalrevolution/2010/07/my-entry-1.html

So you have no preference, as a pedestrian, whether a car that hits you is going 20 or 40?

Or were you thinking we only needed to think about desires of the people in the cars?

When you are travelling in a city, the speed differential to be worried about is the one between your speed and the speed of pedestrians and fixed objects.

Is a street more pleasant to live or shop on if the cars are all going 100 mph vs a mixture of 15 and 20?

Yes, you are right to point out the more general case, but the proposition still holds that speed differential is the problem. If there are pedestrians around the traffic should be slower. I'm just saying that one way streets are not necessarily bad. There is a lot that goes into transit systems and what makes them bad, etc.

One-ways in suburban-ish areas? Baffling. Who even does that?

We have almost entirely one-ways here in Portland ... downtown, where the "moves faster thus more dangerous" factor is irrelevant, since there's a light every block or two.

I'd like to know if the one-way streets had stop signs every 2 or 3 blocks or timed traffic lights so cars could maintain a consistent speed.

Or were they one-mile long one-way streets with no stop signs or traffic lights?

Yes this.

The desirability or undesirability of one-way streets is going to depend on the traffic context. Those streets in Louisville were evidently ones which never should've been one-way to begin with. But can that experience be generalized to other one-way streets? The answer is probably yes for some, but my guess only a minority of them.

And as others have mentioned there are tradeoffs. Efficiency of traffic flow, the environment for pedestrians, the convenience for local businesses, accident rates, property values, etc. There are a lot of neighborhoods whose property values would be enhanced by slower and/or less traffic -- but that doesn't mean that it's a good idea to transform all of a city's streets into slow sleepy suburban 2-way drives.

As for Portland (the one in Oregon), somehow it seems to never have quite upgraded its street system to account for modern automobile traffic. The downtown streets are either one way, or if two-way will typically prohibit lefthand turns. That's probably a necessity if there's not enough room to create a dedicated left turn lane, but I constantly find myself discovering that I'm not permitted to turn left so I have to change lanes to the righthand lane so that I can make three righthand turns instead of the left turn -- and then at least one of those righthand turns will turn out to be impossible due to the one-way streets.

And the inner residential neighborhoods are not well designed for handling even low volume low speed local traffic. The streets are narrow, but parking is typically permitted on both sides of the street, so the streets are physically wide enough for only one vehicle. So I'm constantly turning onto one of those streets, only to be confronted by a driver coming the other way, and we have to quickly and mutually find a space for one of us to pull over so the other person can pass by.

So those streets are terrible for transporting people in a timely manner. The lack of cars and slow speeds does mean that they are better for pedestrians and more importantly (this being Portland) bicyclists. I only have to drive one block on one of those streets to reach an higher volume artery, but that one block is regularly the most annoying and aggravating part of my entire drive, especially if there are 3 or 4 cars trying to negotiate that street at once instead of just two. And if there's a garbage truck or a delivery van ... sometimes I have to literally abort my turn, and put my car into reverse (that's when the presence of the 3rd or 4th car really gums up the works; traffic congestion with just 4 cars).

It's not as bad as driving in Boston (talk about streets that were never designed for modern automobiles), and thankfully drivers in Portland are not jerks as drivers in Boston are, but of the cities that I've lived in Portland's the second most annoying to drive in, after Boston.

Do they allow faster travel?

I live on Beacon Hill in Boston, where most of the streets are one-way. It's awesome for the residents. The layout was clearly planned so that it's very difficult to get from one side of the neighborhood to the other by crossing through it. That, combined with narrow street widths, means that few cars drive on the streets, and people bike and walk down the middle with impunity.

Maybe the Louisville and Beacon Hill examples are consistent. In some neighborhoods, the idea is to discourage outsiders or at least make them go slowly. That's historically desirable for upper income areas because lower, quieter traffic leads to higher property values.. But in the area of Louisville where there are lots of one-way streets, more traffic was a good (e.g. made commercial property more desirable).

I've always found that one way, two lane streets in residential downtown areas lead to more double parking.

Rubbish comment quality on this one, for some reason. Traffic engineering is a fairly mature domain. I would think someone could point to that body of knowledge for some clarity instead of, like, your opinions, man.

I think what you are witnessing is a conflict between "how do we transport people in cars" and "how do we manage the externalities of transporting people in cars?" The mature traffic engineering discipline you refer to is the one that focussed almost entirely on the former, at the expense of the latter, for many decades. Now that is starting to change as cities realize that they can be more than just a place where the road goes through.

This is an excellent comment. Also, the fairly mature domain has recently learned new things.

http://en.wikipedia.org/wiki/Braess%27s_paradox

http://www.wired.com/2014/06/wuwt-traffic-induced-demand/

Love me some traffic. DC does traffic perfectly.

This is a real battle. I am in general concerned regarding Ms. Badger's pieces. My reading of her work indicates she is among those that are opposed to the current level of cars or growth of cars. Now I understand that in all things there should be balance, but when looking at externalities, one needs to look at the externalities of bikes, pedestrians and transit as well as the cars. From that frame work in many cases, I find that cars are not so nefarious and balance may not look as car intolerant as some would propose.

That said, in this case 1325 Brook Street is the Google Photo reference in the Ms. Badger's article. Interstate 65 is 500 feet to the east of Brook Street and Hwy 1020 is 1000 feet to the west, notably a pair of one-way streets. The traffic volume on Brook based on the photos was very low and all of Ms. Badger's comments for this street were right on. It is also true that if these one way pairs are placed in other low traffic volume locations returning to two way streets is a very good and valuable change for both residents and travelers.

However, Hwy 1020 was conveying significant traffic and two way (by street) operations would substantially reduce the capacity of the one-way pair and disperse the concentrated traffic throughout the neighborhood grid of streets. I would predict that congestion on Hwy 1020 would go up if converted back to two-way streets. I would predict that traffic accident off Hwy 1020 would go up and property values would go down both on Hwy 1020 due to the increase congestion and property values would go down in the adjacent neighborhood due to the diverted through traffic.

Context is key, One-way streets on low volume roads are likely not efficient, One-way pairs that convey travelers along Hwy routes such as Hwy 1020 may be better than the alternatives such as 1) Acquiring ROW and widening 2nd street so it handles the traffic on one street; 2) Disbanding the formal Hwy 1020 and allowing traffic to flow anywhere in the grid.

I would not suggest the experiment, but applying the two one-pairs theory to Hwy 1020 would be a super productive economic experiment, but maybe not so good for Louisville.

Comments for this post are closed