Avoid Bikelash By Building More Bike Lanes

Market Street, San Francisco.

pfb logo 100x22Michael Andersen blogs for The Green Lane Project, a PeopleForBikes program that helps U.S. cities build better bike lanes to create low-stress streets.

Here’s one reason the modern biking boom is great for everyone: more bicycle trips mean fewer car trips, which can mean less congestion for people in cars and buses.

But there’s a catch. A recent study shows that when bicycle use rises but cities don’t add bike lanes to put the new bikers in, traffic congestion actually gets worse.

In some situations, it gets a lot worse.

A study measured travel delay on a street with bikes but no bike lanes

NE 47th Avenue, Portland.


It’s happened to most regular bike users; it happened to me last week. Biking to meet friends at a restaurant, I had to pedal two blocks uphill on a street without bike lanes. As I started to push up the slope, a man zoomed his car around me, straddling the two lanes and laying on his horn as if I’d done something wrong.

I’d love to be out of your way too, I wanted to tell him. But this parking lane would have to go.

I had every legal right to use the traffic lane. But a 2013 study from the University of Virginia’s civil engineering school shows that in a sense, the man wasn’t wrong. I was indeed slowing the whole system down.

“Safety, comfort and all that aside, the efficiency of a dedicated facility is that you can maintain varying speeds,” said Alec Gosse, an author of the UVA study, in an interview this week. “If you’re on some sort of major street with bus traffic in particular, then you end up leapfrogging the bus. And no one is happy about that.”

If you don’t add bike lanes, each additional bike makes car and bus travel slower

The model studied streets with one travel lane in each direction and a bus line that could get stuck behind a bike.


The study, co-authored by Gosse and his professor Andres Clarens, used a traffic simulation to compare four theoretical two-way, two-lane streets:

A) One where bicycles were 1 percent of traffic (about the level of many U.S. cities) with no bike lanes.

B) One where bicycles were 1 percent of traffic with bike lanes.

C) One where bicycles were 10 percent of traffic (about the level of Eugene, Oregon, or Boulder, Colorado) with no bike lanes.

D) One where bicycles were 10 percent of traffic with bike lanes.

The street with the most congestion? It wasn’t Street A, which represents so many in the United States today.

The most congested street was Street C.

Even after Gosse and Clarens assumed that cars, trucks and buses would have a chance to pass a bike every 100 meters or so, the increased biking only slowed everyone down.

Jonathan Lewis, deputy director of transportation in Atlanta, said he’s noticed this in action.

“The phenomenon is real,” he said. “We have some busy streets that definitely are nowhere near 10 percent of the volume, but there’ll be a few bikers on there every rush hour. And the impact is huge. So I think if a street is near capacity and you have a few brave souls trying to bike on it, then they can have a pretty big impact. And then they don’t want to sit at the traffic light, so they’ll jump the whole queue and start the whole process over again.”

The problem seems intractable. Even if a city succeeds in boosting bike use, it faces a backlash once the new riders start slowing traffic.

Fortunately, the study by Gosse and Clarens also suggested a way out.

If you do add bike lanes, each additional bike helps everybody

10th Street, Atlanta.

To be sure, travel speed isn’t the most important thing in the world. On some streets, a little travel delay can be great for both local businesses and driver safety.

And the UVA findings don’t mean that people shouldn’t be allowed to bike in travel lanes on popular roads. Restricting public streets to people who own cars would be both unjust and impossible to enforce.

But the study does mean something very important to the politics of bike infrastructure.

If a city doesn’t build bike lanes, then “bikes vs. cars” is actually real. But if a city builds bike lanes, more biking becomes a win-win.

Public support for bike infrastructure and programming depends on one crucial concept: that more biking benefits people whether or not they ever ride a bike themselves. Lewis, the Atlanta transportation deputy director, said that’s one big reason it’s important to add bike lanes to busy streets when possible.

“It helps people not hate bikes,” he said. “If you’ve got the space, of course the bike lanes will help congestion.”

The model studied streets with one travel lane in each direction and a bus line that could get stuck behind a bike.


But for a city to keep its bike lane network growing, where will the road space come from?

On streets with more than one lane in each direction, the best answer is often a road redesign that also adds a center turn lane or removes passing lanes. But for two-lane streets, the UVA study offers another useful answer: curbside parking spaces.

Because the Gosse-Clarens model calculates the anti-congestion benefits of a bike lane to the transportation system, it offers a way for cities to calculate how much money a curbside parking space needs to be worth in order to justify its existence.

“If you have a parking space, that’s basically going to provide a service for, I don’t know, five, eight people for day?” Clarens said. “I like street parking as much as the next guy. … But at the end of the day when you think about what is the total societal cost of that pavement, it can be allocated better; it can be used better.”

Lewis said that’s exactly what Atlanta has concluded, too. It’s now working on several projects that will replace parallel parking with bike lanes.

“That’s been an important opportunity for us,” Lewis said. “It’s just a better use of the pavement.”

You can follow The Green Lane Project on LinkedIn, Twitter and Facebook or sign up for its weekly news digest about protected bike lanes.

41 thoughts on Avoid Bikelash By Building More Bike Lanes

  1. What I think many people who drive don’t realize is that, by not sharing the road with slow moving cyclists, they are creating the need for separate bikeways, which usually means the removal of a travel lane or parking. Many drivers who complain about parking loss or lane removals should be mindful that these types of projects are primarily because of the danger and discomfort they produce.

    Ultimately though, once people get used to whatever change happens on a street that gets a bikeway, they appreciate that everyone has their own space which makes it much more predictable where everyone will be.

  2. This is probably why many recent road diets/complete street transformations in the US have typically seen a reduction in conflicts and injuries amongst users for *all* modes–whether you’re in a car, on foot, bike, wheelchair, or whatever.

  3. This wasn’t covered in the study, but I have noticed that the investment in bike lanes has resulted in a lot fewer bicyclists using the sidewalks.

  4. Which is why I support investment in bicycle infrastructure, even thought I don’t ride a bike!

  5. Note this was done with a traffic simulator where the answer really depends on your assumptions. Also note the baseline speed of traffic on the road is rather high (approximately 65 kph). This is not representative of city streets, where congestion is already too high to support this sort of average.

  6. Yup, exactly. Good road diets/complete streets really benefit all users. The ensuing reduction in injuries and conflicts amongst all modes is really a major win for everyone.

    In addition to the benefits to people on foot due to reduced crossing distances (and the sheer buffer effect that protected bike lanes provide from cars), protected bike lanes also directly benefit other non-bike modes who can also use a cycletrack (rollerblades, skateboards, mobility scooters, etc.).

    This is especially relevant for the less abled:

  7. Those simulated cyclists need to read my book and take my class to learn to be vehicular simulated cyclists and ride at the speed of traffic. After just six 2-hour sim-u-cast classes, those simulated cyclists will be able to ride with simulated traffic at ANY simulated speed!

  8. As someone who is put on both sides of this dynamic, I can very much relate. As a driver in a city that is seeing increased biking but virtual no infrastructure, or no infrastructure that meaningfully separates and protects bikers, it can make drives a very frustrating exeprience. It’s a very nervous, uncomfortable, and sometimes dangerous experience whenever I’m on the other side of the equation.

    Although I know there are still some around these parts that believe otherwise, I would say “no, we can’t all get along.” What bikers and drivers need is often fundamentally mutually exclusive and no amount of unrealistic lawful/behavioral perfectionism is going to change that.

  9. Hopefully our streets can at least be safe enough that when some newbie comes along and rides on the wrong street, they are safe enough that they’re able to live long enough to learn which streets are better.

    And the poor cyclist that is taking their first or last leg of a trip and has to leave a decent street for one that’s much less safe.

  10. Yes, exactly. Above certain speed thresholds–this applies especially to collector roads and arterials–there’s every reason to not mix such disparate modes as cars and bikes whenever possible. Cars are not bikes are not feet.

    The other benefit to drivers is that the more attractive a street is to walking/biking/transit, the fewer people hop in their cars anyway = less competition drivers face with other drivers. For example, imagine if there were no transit or bike lanes or sidewalks in the following video. Imagine every person you see in this scene replaced with a single-occupant, separate car:

    It’d be car-traffic pandemonium. Yet as you can see for the few cars using this road (in the distance) it’s clear and scot-free for them and everyone gets along as it’s clear where everyone goes.

  11. A lot of drivers use the bike lanes as double parking lanes, and the useless SFPD doesn’t do a thing about it (especially on Valencia St.), so I believe drivers view the bike lanes as a bonus for themselves as well.

    The bike lanes/double parking lanes seem to also keep motorists from the previous tradition of parking on the sidewalks which at least is helping to prevent pedestrians from the risk of walking out into the streets.

    From my cyclist’s view, I guess the bike lanes/double parking lanes are better than nothing, but they don’t really seem safer to me considering the double parkers force us cyclists to swerve into motor vehicle traffic. Another safety concern is that a lot of the motorists don’t seem to know they’re supposed to merge into the bike lane prior to making a right turn at intersections. That combined with the modern lack of use of blinkers for lane changes & turns makes the bike lanes not as safe for cyclists as they should be.

    At this point, I really don’t care anymore about the bike lanes and don’t care if more are added or if they’re removed. I’m fed up & sick of most of the people in this city these days.

  12. A lot of drivers use the bike lanes as double parking lanes, and the useless SFPD doesn’t do a thing about it (especially on Valencia St.), so I believe drivers view the bike lanes as a bonus for themselves as well.

    Absolutely. That’s where parking-protected bike lanes should come in:


    Imagine if this continued up all of Polk. Or existed anywhere on Valencia.

  13. As Mark Twain has been attribute to have said, “There are lies, damned lies, and statistics.”

    Let’s not play so fast and loose with these numbers.

    You show me a graph with very little resolution and you make me guess what the actual numbers are. When 1% of trips are by bike, there is little, actually no discernible difference between bike lanes or not. They both appear to be at 55 seconds/kilometer.

    But if bike usage rises to 10% of all trips … unusual in most of the US, but this is your game, so I will play along … the no bike lane time of travel is (I think) 87.5 seconds/kilometer without bike lanes and, shazam, exactly the same as the 1% usage time with or without, 55 seconds/kilometer.Amazing. Your number fit your thesis. I distrust numbers that work out so neatly. Did the people doing this study ever work at University of East Anglia?

    The difference between the two is about 32 second/kilometer. That sounds quite significant. Since you and I are both in the US and what the heck, even in the UK they use imperial on the roads (why is that? If you’re going to go metric, go all the way) that is a delay in a 10% bike usage environment of about 51 seconds/mile.

    Gosh that sounds like a lot. But the study doesn’t account for the fact that a car driver rarely follows a bike for a very long period of time. In my own real world (not computer simulated) experience, I have never had a car dwell behind me for more than one minute. Almost always, they pass me before a 15 seconds have elapsed. They never travel behind me for a full mile, much less a kilometer. And it takes about 5 seconds for a car to pass a bicycles with a 15 mph speed differential. The creators of this study even say as much, “Realistic urban corridors vary in width, so it is overly conservative to assume that a given lane width will restrict passing movements indefinitely.” Translation, “In the real world, you will get a chance to pass very soon, be patient. But for the sake of pushing an agenda of herding cyclists into narrow, poorly maintained gutter lanes and getting them the hell off our car lanes, we will assume that the car driver will be stuck behind a cyclist and a bus FOREVERRRR.”

    What makes this study particularly flawed is that it uses a computer simulation and doesn’t take into account real world conditions. I can, and in my past I have, tweaked a program used for simulation so my bosses got the results they desired.

  14. The problem with bike lanes is that the City Leaders want to court that demographic that wants bike lanes. Unfortunately, doing it right is expensive. The largest part of the demographic that wants bike lanes want them to get “those bikers out of my lane” but they aren’t ready to pony up what it costs to do it right. Extensive networks of protected bike lanes cost lots of money, tax money, and “steal” “car” lanes or parking lanes, or both. So what you get are either extensive networks of painting projects or you get short bursts of good intentions that go from no where to no place. In either case, there is rarely any money budgeted for proper maintenance or even snow removal. Those paint project are not only not budgeted for snow removal, they are used as a place to put the removed snow.

    Until our cities are willing to spend the money to do it right and maintain what they do, I would far rather see that money be used to educate driver about how to behave around bicycles and how driving is a hard thing that carries a huge amount of responsibility.

  15. I don’t have much time to read this study beyond the abstract and again, I’m not really buying any of this. Again computer simulations using a scenario set to maximize the potential delay of cyclists riding in mixed traffic. No places for cyclists to move over to let traffic pass, no considerations for the reality of traffic lights stopping traffic or the reality that cyclists are likely to take alternate routes where they exists and are very convenient for the authors.

    AND is it so bad to slow down traffic to a more human speed anyway?!?! How many times have we had drivers fly by at 50mph only to get stopped at a traffic light which turns green the moment the bicyclist arrives?!?!

    Again, not against well designed bike lanes and protected cycletracks but this study is not really based on reality.

    PS – Do you always have to show some of the worst bike lane designs out there?!?! Two-way cycletracks are almost always done as a compromise that trades cyclists real safety for perceived. Market Street however looks great!

  16. Yeah, those bollards are a joke.

    I’ve been wondering if there’s a hollow in the asphalt at the base of those bollards because I’ve been considering putting concrete inside them, maybe one at a random time, but I think some kind of support to keep them from flopping over before the concrete sets would be required.

  17. I guess we were lucky to get the magic paint rather than none, but it’s the wrong paint. We cyclists need the kind of magic paint that wards off evil motor vehicles, not this kind that attracts motor vehicles and makes bike tires slide like ice skates in the wet weather. The witch who cleansed & charged the magic paint must be a motorist who’s playing games with us or made a mistake.

  18. I think those small protective concrete plots are intended to serve as burial plots for cyclists who get run over on the rest of the unprotected parts of Fell & Oak streets.

  19. Are cars driving in Market’s bike lane? If so then yes you got a point. If not, I don’t know how concrete will help.

    I was in San Francisco a month ago and rode through the western side of town and liked what I saw. I found the facilities useful and reduced my traffic stress even as a more experienced cyclist.

  20. I feel much safer with security bollards that have the capability of stopping an “out of control” car. Amazing how my bike never goes “out of control.”

    Like in these hilarious videos of cars trying to “beat the bollard.” Hint for moron car drivers: the bollard always wins. See:


  21. Although I really do not like to do so, I myself have been known to ride on a sidewalk when the alternative is playing tag with two-ton lethal weapons.

  22. Biking down a one way street, a driver from the rear honks at me, passes me and comes to a stop at the intersection for a red. He rolls his window down, however I ignore interacting with him. As we both make left turns he twice tries to force me out of the left turn lane. He proceeds to yell, ” hey buddy you cannot be biking wherever you want.” At a right turn in the next block, I have a bike lane, but driver drives alongside me and starts yelling a-hole. Not sure what his argument was there.

  23. The Polk Street debacle show that the MTA crumples like a wet paper towel at the slightest sign of disagreement.

  24. this is a ludicrous study. the addition of bike lanes is clearly causing mroe congestion in SF by taking away from the roadwway. there is no biking boom. In fact, desptie the plethora of new lanes, cycle commuting in SF is still below 3.5%. Most of the bike lanes we have are empty the vast majority of the time . why would we add more.? waste of money and boondoggle catering to the elitist white male under 45 that make up <3.5% of commuter all at the expense of the 96.5% who use the roadways via bus, taxi, private auto, uber, etc.. How this tiny minority has overtaken city policy is beyond me

  25. Perhaps you’re just upset that for most areas of San Francisco, cycling is the fastest way to commute.

  26. Except perhaps in cities such as Palm Springs, where sidewalks are often the designated bike lanes…

  27. this is a ludicrous study. the addition of bike lanes is clearly causing mroe congestion in SF by taking away from the roadwway

    Crazy bikes, taking up all that room on the road. Let’s replace them with more vehicles to finally end congestion!


    Most of the bike lanes we have are empty the vast majority of the time . why would we add more.?

    When protected bike lanes like the above are built, people use them. It’s the subpar ones people avoid. Lesson learned…bad bike lanes are bad. So let’s stop building them:


    I wonder why no one’s using the bike lane!

    there is no biking boom.

    In fact, desptie the plethora of new lanes, cycle commuting in SF is still below 3.5%.

    –> Biking to work is not the same as bike modeshare
    –> Nor is biking to work the only or even primary low-hanging fruit.
    –> It’s the other trips we make throughout the week of 1-3 miles or less.

    That being said, SF’s modeshare goals of 20% are indeed not being backed up by infrastructure that would support such a goal. C.f. any double-parking/doorzone bike lane.

    waste of money and boondoggle


    –> 0.46% is pretty cheap for modeshare that’s currently at least 10x that.

    –> Modeshare goal is 20%

    elitist white male under 45








    But yes, let’s replace all the bikes with more cars. To end congestion. SF definitely has the room for that.

  28. What percentage of city roadway is dedicated to cycling vs. what percentage is dedicated to automobiles? It is at least 100:1. We should get at the very least our 3.5%. It is not my fault you chose to commute in a large inefficient vehicle that takes up an extremely oversize amount of space.

  29. Yup.

    –>Parking lanes in San Francisco constitute 15 percent of the paved roadway area, equal to real estate valued between $8 and $35 billion.

    –> Street parking in San Francisco totals 902 miles in length, six times longer than the 143 miles of bike lanes.

    –> 75 percent of all bike lane miles were built since 2000.

    –> Bicycling constitutes four percent of trips, but only 1.4 percent of roadway space is dedicated to bicycle lanes.

    –> There are 36 lane miles of dedicated transit lanes, but 211 lanes miles of freeway lanes.

    –> General tax revenues, not user fees, pay 75 percent of roadway maintenance costs in San Francisco.

    –> The federal gas tax, in inflation-adjusted dollars, is at its lowest point since 1983, when the Reagan administration doubled it.


  30. @Jimbo – Where’s your non-ludicrous study, the one that substantiates your claim about what is “clearly causing [more] congestion?” Perhaps you could be a little more specific about “clearly” means in this context.

Leave a Reply

Your email address will not be published. Required fields are marked *


Surprise! People Aged 60-79 Are Leading the Biking Boom

Michael Andersen blogs for The Green Lane Project, a PeopleForBikes program that helps U.S. cities build better bike lanes to create low-stress streets. The national surge in adult bicycling since 1995 may have more to do with healthy hips than with hipsters. Biking rates among people between the ages of 60 and 79 are soaring, […]

Why Do African Americans Tend to Bike Less?

Michael Andersen blogs for The Green Lane Project, a PeopleForBikes program that helps U.S. cities build better bike lanes to create low-stress streets. It took a week in Copenhagen for Albus Brooks to start thinking seriously about bicycling. The Denver City Council member, 35, had never owned a bike. By the time he headed home […]

Bike Lanes Don’t Lead to Congestion, But Some of Them Should

Gretchen Johnson and Aaron Johnson have posted a nice debunking of typical “war on cars” rhetoric over at fivethirtyeight. Johnson and Johnson gathered before-and-after traffic data from 45 miles of streets where Minneapolis installed bike lanes. They also looked at how Brooklyn’s Prospect Park West bike lane affected traffic conditions. They found, in short, that after the installation of […]