London’s Protected Bike Lanes Move People 5 Times More Efficiently Than Car Lanes

A 2014 artist's rendering of Blackfriars Junction, immediately west of Upper Thames Street, after protected bike lane installation.
A 2014 artist's rendering of Blackfriars Junction, immediately west of Upper Thames Street, after protected bike lane installation.

PlacesForBikes is a PeopleForBikes program to help U.S. communities build better biking, faster. You can follow them on LinkedIn, Twitter or Facebook or sign up for their weekly news digest about building all-ages biking networks.

Leave it to a Brit to deliver a mathematical smackdown this courteously.

History may never record which anonymous bureaucrat was assigned to field the following question about London’s protected bike lane network (known there as “cycle superhighways”) submitted to the public agency Transport for London:

Prior to the introduction of cycle superhighway, certain claims were made by TfL on the impact on traffic on Upper Thames St. Congestion now seems to be worse than predicted. Please supply any data or reports on the prediction v. reality. If no analysis has been done, please let me know if it will be and if not, why not. Thank you.

The bureaucrat found while looking up the data for this question that some auto speeds (especially the eastbound evening rush hour) were indeed slower than first modeled, while others (the westbound morning rush hour — so, in most cases the same commuters) came out faster than first modeled:

That’s when the Transport for London bureaucrat also dropped this mind bomb into the conversation:

Recent monitoring data shows that central London segregated cycle lanes are moving five times more people per square metre than the main carriageway, with East West Cycle Superhighway seeing a more than 50 per cent increase the total mileage cycled.

Because the person who asked the question, after all, didn’t ask about auto congestion. They asked about traffic congestion.

And the happy news from this project seems to be that because “traffic” means people, not cars, London’s new protected bike lane network has made Upper Thames Street much, much more efficient — because it gave more Londoners a way to comfortably move through their city in a way that requires less public space.

Via Last Not Lost.

33 thoughts on London’s Protected Bike Lanes Move People 5 Times More Efficiently Than Car Lanes

  1. As some of your readers know, I was for many years transport correspondent of the Financial Times, based in London. I was the FT’s US industry correspondent, based in New York, for four years. I used to write a blog, the Invisible Visible Man, about my cycling experiences. I recently stopped writing it because I was getting so much grief from London’s cycling community for pointing out the inaccuracy of much of the cycling community’s thinking about these cycle lanes. I support the building of the lanes but recognise they’ve increased motor vehicle congestion. I’m disappointed to see you’ve posted a story reinforcing the magical thinking of many in the cycling community and some in TfL about their impact.

    The biggest problem with TfL’s response is that it doesn’t compare the current journey times with past journey times but against “modelled journey times”. Those are the expected journey times after construction, rather than the real journey times before. TfL’s previous impact reports have made it clear that the north-south journey times on the north-south cycle superhighway were five to seven minutes before construction, not the “modelled” 12 to 19 minute journey times. The current journey times are, unsurprisingly, all longer than the journey times before several lanes of traffic were taken away.

    Meanwhile, you praise TfL’s line that the roads move more people at peak hours than they did before the cycle lanes were put in place. This is a very partial view of the situation. The unique capability of London’s roads is not that they move people but that they move lots of other things. So there’s more to the slowing down of road traffic than the slowing down of people. It matters that deliveries of building materials, packages and so on are caught up in ever slower-moving traffic. There’s little scope for moving motorists onto bikes in central London because only 20 per cent of the traffic is private cars. People are transferring to bikes largely from public transport. So it’s fallacious to suggest the bike routes are facilitating a shift from cars to bikes. There aren’t enough private cars left for that to be a big factor.

    Also, TfL carefully chose only the peak hour. Because London’s roads are full of motor traffic, the peaks are much longer than an hour but the peaks on the superhighways are shorter. It’s clear to any looking at the situation that, say, at 4pm, the motor sections of places like Blackfriars Bridge are full while the cycle sections are lightly used.

    I use the cycle superhighways daily and I’m delighted they were built. It’s vital, however, that people stop fooling themselves about the impact of taking away significant road space on a road system that was already badly congested and becoming more so. TfL projected big increases in journey times after construction. Some results have been better than those projections, some worse. We should stop pretending otherwise.

  2. The planners won’t and can’t say it outright for political reasons (and I don’t blame them) but I will:

    The idea is to make using a private car so unbearable that people will stop doing it.

    Removing traffic lanes to make cycle tracks is not only to encourage cycling but to discourage driving.

    A lot of people will see this as authoritarian but those are also the same people that think driving is a right rather than a privilege.

  3. Unlike in many US cities, TfL want to get people off public transport, especially the tube, but also major bus routes. Seeing the actual throughput figures would be good, and would certainly show that this nascent network is delivering major benefits. What is Robert Wright’s alternative? TfL is actively looking at major restrictions on private cars (for air quality reasons) wouldn’t a 20% reduction (% private cars) be good?

  4. It certainly wasn’t the stated purpose of the cycle tracks’ building to do that and I hope it wasn’t the intention in central London. That’s because the private car is pretty much a thing of the past inside London’s congestion-charging zone. They account for only around 20 per cent of traffic. My concern is that buses, deliveries of all kinds, trucks carrying building materials and so on are getting caught up in this. That’s a serious problem.

  5. If you bicycled, you would know that the intention of cycle tracks is to protect the safety and the lives of bicyclists.

    Try getting on a bicycle and riding in mixed traffic, and your windshield perspective will change.

  6. A reduction in private vehicle volumes would be good. But the 20 per cent of road users that are in private cars in central London now are a pretty hard core bunch. They’ve put up with a significant congestion charge and significant congestion. There must be disproportionate numbers of disabled people, people who feel some security need to be in a vehicle or whatever. I don’t know where you get the idea TfL wants to get people off public transport. The last time I spoke to their head of surface transport, he was unhappy about the decline in bus patronage as a result of congestion. It’s pushing up the costs of bus operations for them. Nor are the new restrictions on private cars “major”. They’re introducing a charge for highly polluting vehicles. But it will catch only a small proportion of existing vehicles.

    The whole reason I am unhappy about the misrepresentation of this issue is that I’m unhappy it’s distracting people from the urgent need for a more sophisticated congestion charge. I wrote about that in my (former) day job here; I mentioned it in a (private) blogpost about this issue here:

  7. 20% is a lot and taxis/rideshare are almost as bad as private autos so perhaps those should be culled to make way for buses. Trucks could be reduced in size also and lots of deliveries could be done by bike.

  8. I think you misinterpreted my position. I’ve ridden about 7000 miles this year, I’m all for cycle tracks (even though I prefer to ride in the road when I’m going fast) and making the barriers to riding lower.

  9. I sat for nearly an hour in congestion in a London bus last year on New Cross Road. Walking would have been faster.

  10. It’s true. About 50% of those advocates who argue for better bike and bus infrastructure actually want to make things better for bikes and buses. The other 50% just want to stick it to cars regardless. The former are more credible.

  11. ‘20% of road users are in private cars’ is a very, very deceptive statistic. If a bus is holding 100 people then there are 20 cars on the road for that single bus? A better statistic would indeed be percentage of vehicles, no?

    If I where to level accusations of misrepresentation, I would be meticulous in my argumentation to avoid my own misrepresentations

  12. Just getting through the rest of your comments…. if your concern is really about the movement of people and goods through central London, then if ywe were to remove those 20% of commuters using their private cars and put them into buses (bikes neither here nor there – mind), then ALL road users including those indisputably important goods and buses would flow much more quickly.

    We can agree that we are dealing with using a very limited resource as efficiently as possible to the greatest benefit of all. Cycling accomplishes this, mass transit accomplishes this. Wide, safe pavements accomplishes this. The private car destroys this.

    That would choose to take a stand on this single point and throw up your hands, ‘there is nothing you can do about it!’ leaves me a little lost.

    I am not currently a resident but have been and am there often to visit family and on business. My experience is that the ease of transit across the city has improved remarkably over the past 15 years.

  13. This is fair criticism, but (as the author) it seems to me that this piece also implies that bus passengers are traffic and should also get dedicated road space. Isn’t their congestion the fault of the inefficient mixed traffic lanes, not the efficient bike lanes?

  14. Truly, one of the least attractive things about cycling activism at present is the prevalent smart-aleckiness of much of the commentary. Transport for London put out a deliberately misleading statistic that is being (wrongly) trumpeted in this piece as a work of perspicacious genius. I used a slightly infelicitous term to describe something and you knew very well what I meant. But, yes, make out I’m the problem here.
    London’s bus users are being done over at the moment because the buses are stuck in ever slower-moving traffic. I’ve been careful to advocate for their interests when many of my fellow cyclists have assured me it’s a good thing that a transport mode used daily by 6.5m Londoners, many of them the capital’s poorest people, is suffering because the vehicles mostly have internal combustion engines. I’d like to see people who care about these issues getting worked up about that problem instead of picking holes in people’s phrasing.

  15. I’m really not sure what this comment means. All the data suggest that congestion for motor vehicles of all kinds in central London has grown significantly worse since the building of the new cycle facilities. While London has a lot of dedicated bus lanes, the growing congestion on the rest of the network has inevitably held up buses and driven passengers away, many of them to Uber. It doesn’t make any sense to say that it’s not the fault of the cycle facilities because it inescapably is. I support the building of the cycle facilities. I’ve used the east-west cycle superhighway and the north-south superhighway already today and I’ll be using the north-south superhighway again this evening. But there’s an unfortunate tendency, typified by the disingenuous FOI reply that inspired this piece, to pretend that taking away lots of road space from major roads in a city where congestion was already worsening is somehow unrelated to the significant worsening of congestion that’s followed. When the motoring lobby deploys arguments so obviously misleading, we go crazy. I don’t see why we have to smile and nod when the misleading nonsense is being put out by our own side.

  16. This is correct. But it doesn’t say anything useful about this situation. This is about the removal of motor vehicle lanes in an already severely congested city to provide cycle paths. While the cycle paths are excellent, they are mostly lightly used outside peak hours, while the motor vehicle lanes are congested for much of the day. So, despite the misleading comparison in this FOI, it is clear that motor vehicle congestion has grown worse as a result of the new lanes’ installation, even though motor vehicle volumes have fallen.

  17. I’m not asking anyone to smile and nod at nonsense, and I appreciate your local knowledge here – obviously it’s greater than mine. But I don’t think I understand your point either.

    Are you questioning TfL’s claim that the superhighways carry more traffic per square meter?

    Even if not, I can see the case that this ignores lorry traffic, but is there a strong case for prioritizing lorry traffic in the peak hour anyway?

    As for buses losing passengers to Uber, isn’t that best corrected with more bus priority and less priority for 2-occupancy cars?

    The reason the 5x factoid is significant is that whatever their proximate impact on auto speeds, converting auto space to bike space was a net gain for the total capacity of the transport system during peak hours. Low-occupancy motor vehicles are the least efficient use of road space in the peak hour, so any congestion reduction should involve a mode shift away from them. No?

  18. The claim that the bike lanes carry more traffic per square metre isn’t, I think, meaningful in any real sense. It’s a claim that they’ve been making in one form or another ever since these facilities were built and it doesn’t capture the reality of what’s happening.
    The claim was originally made about Blackfriars Bridge, where TfL said the bridge now moved more people in peak hour than it did before the bike paths were put in. The bike path on the bridge is very busy at peak times and it does at those times move large numbers of people.

    But this claim – and the similar claim they’re now making about the east-west cycle superhighway – ignores nearly everything useful to know about how traffic actually flows in central London.

    First of all, the peaks on the bike paths, which are used almost exclusively for commuting, are very short, while the peaks on the motor lanes last practically all day. So, while Blackfriars Bridge moves a lot of people in peak hour because the bike path is full, for most of the day the bike path isn’t anything like full while the motor lanes are. I can only imagine that the claim about per-metre throughput on the east-west cycle superhighway applies to the peak, though I note they don’t say so.

    Secondly, this cannot be a case of the virtuous transfer of private vehicle trips to bicycle that frees up traffic. Only 18 per cent of the driving in central London is by private cars, according to table 6.6 here: While another 32 per cent are taxis and private-hire vehicles, an awful lot of the rest of the traffic is made up of vans or trucks of various kinds. Counting the number of people in these vehicles doesn’t make much sense to me. Nor does it make sense to suggest, as you do, that somehow vans and trucks don’t belong on the roads at peak times. There aren’t quiet times during the day when they can move – the roads are full all day. They could move at night but there are bans on truck movements overnight, so that residents living near shops don’t get woken up early in the morning. Any claim that the installation of the superhighways represents a net gain for the system in these circumstances relies on the idea that the journeys being made would all have been made on road before. In fact, I’m confident most cyclists are coming from the underground or surface rail (albeit some must also be coming from buses). There are very few private car trips to transfer. It’s not self-evident that a substantial slowing of motor traffic that’s facilitated a shift from underground and rail to bike has improved the system’s capacity overall.

    But the biggest problem with the claim about the different throughputs is that it probably says far more about the slow movement now on the motor lanes in question than it does anything positive about the bike facilities. When I ride along the east-west cycle superhighway, as I do nearly daily, I pass, at nearly all times of day, a nearly stationary line of traffic. TfL’s impact report on the cycle superhighways is clear that traffic used to move far faster along there. The slower this traffic moves, the easier it is for the bike path to beat it in square-metre throughput. The logical answer to this line of thinking is that it would be better to hand another lane over exclusively to cyclists so that only, say, five people per hour could get along the motor vehicle lanes. If six people per hour used a similar space of cycle lanes, TfL could wave a flag and congratulate itself on its triumph. I don’t think this is the optimum way to do things.

    I’m attaching a picture of the traffic outside my office this lunchtime as an illustration of the issues here.

    I can only stress again that I am an enthusiastic advocate of cycling and a regular and keen user of these facilities. I am simply wary of misleading arguments from TfL that don’t reflect the reality as described in the wider statistics. I want more segregated bike paths but I think it’s vital that this is done in the context of a sensible transport policy that will, for example, ensure that the material for building the new bike facilities doesn’t spend long periods waiting in traffic jams. Roads are economically vital facilities and we shouldn’t let ourselves believe convenient myths about them.

  19. Hey Robert – I saw a comment you submitted in response to this in my email. Don’t know what happened to it but I wanted to say that I think you make a pretty good case about bike lane space being underused during off-peak hours. I haven’t been to London in 16 years (so, since before the decongestion charge began) and I suppose I may not be appreciating the extent to which off-peak congestion is a major problem on surface streets.

    Also, your observation that bike trips are combining with rail trips squares with the data I’ve seen that most of London’s new bike trips were previously mass transport trips. There’s something to be said for making the Tube more attractive and useful by taking pressure off it, but that’s no guarantee of moving people out of cars.

    As for bike lane “efficiency” being due to their ability to flow freely during peak hours, I think that could be framed as a feature, not a bug … but I think you’re right that sometimes it’s relevant to compare vehicle speeds too.

    Ultimately I find it hard to imagine that Amsterdam isn’t the most space-efficient direction for London to move its transportation system. Excellent/plentiful bike infrastructure is an essential ingredient of that.

  20. Almost every progressive large city is slowly trying to do this. London is a bit further down the road…heh

  21. I wonder what the movement towards bikes will have on family diversity in these areas, primary in terms of family size.

  22. Ok point taken, now about the statistic you put forward about road user percentages? I don’t think it to be a smart ass at all to call you out on that – asserting that it is deceptive.

    But sure, take down your blog, quit your reporting and instead post on an advocacy site because of the ‘abuse’ you’re getting. If I am representative of it, then methinks you need a cheering section and not an honest exchange.

  23. My logo image is the bicycle I pedaled across the USA in 2000. Where I could I used lengthy bike paths. They were the safest, as were, when allowed, the full lane width shoulders of the Interstates. Secondary two lane highways from the first era of higher speed automobile highway building were much less safe to ride, and I remarked that I had exercised my “death wish” enough for one year to a reporter at the end of my trip.
    One of the more pleasant separated bikeways utilized the former route of the Erie Canal towpath. It had been mule speed at inception, then converted to rail, and with my aerodynamic bicycle I probably traversed it as fast as those trains used to go.

  24. Hi Robert,

    The E-W route along the embankment was specifically chosen because there are so FEW bus movements along it.

    The alternative was to run the route through The Strand, Fleet Street, Ludgate Hill & Canon Streeet and that has 20x more bus services an hour.

  25. Hang on… The evidence is that introducing an additional 60,000 Ubers into the city has impacted congestion. That should be the focus not reducing active travel options.

  26. Do you have data that indicates the cycle paths are used outside of peak hours, or are you just assuming that? In my experience in other places I have lived (the US, and the Netherlands) the peak hours for the bikes and the cars is largely the same, because people are commuting to work or school. I personally don’t have data for this on the region being examined above, but I find your assumption questionable based on what we are offered here.

    Also, do you have data indicating that the motor vehicle volumes had fallen? Because I didn’t see that indicated in this report. Increasing congestion is to be expected in areas with population and job growth, even when no lane changes are made, so citing the bike lanes as the cause of the congestion isn’t as straight forward as you are making it seem.

    If overall volumes of people are being moved through when you add cycles and motor vehicles together, which is what this report seems to be indicating, though didn’t give us the exact numbers, we are seeing an increase in efficiency overall, which would indicate that cycle lanes are more than making up for whatever impact they had on motor vehicle travel time.

    I personally, will need to see more information to support your base assumptions before I can get on board with your logic here.

  27. London has the density to make this practical and attractive. Development policies can do a great deal to create density, such as zoning for low-impact pocket retail in walking/cycling distance to people in nearly neighborhoods.

Leave a Reply

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


Cycling Booms in London, and the City’s Not Looking Back

Boris Johnson says that one of his goals as mayor of London was to make cycling “more popular and more normal.” As Johnson’s eight-year tenure winds down, it looks like the progress he made in his second term has accomplished that mission. If current trends continue, bike commuters will outnumber car commuters in central London by 2018, according to a […]

Cycling Up 70 Percent on London’s Bike Superhighways

Figures are starting to come in from London’s investment in “cycle superhighways,” and the data make a strong argument for more robust dedicated bike infrastructure. A Transport for London study found that cycling is up 70 percent along routes where the city’s beefed-up bikeways were installed, reports Network blog Part of the increase is […]

Cycling Up 70 Percent on London's Bike Superhighways

Figures are starting to come in from London’s investment in “cycle superhighways,” and the data make a strong argument for more robust dedicated bike infrastructure. A Transport for London study found that cycling is up 70 percent along routes where the city’s beefed-up bikeways were installed, reports Network blog Part of the increase is […]