Friday, 13 July 2012

In which I "dangerise" cycling again

On Thursday I attended the session of the London Assembly Transport Committee on cycle safety. Some of the things I and another Cycling Embassy of Great Britain member, Tim Lennon, said in the debate have been reported here. I won't give a general account of the debate, in which representatives of London Cycling Campaign, British Cycling, CTC and Sustrans all said some very good things which I hope will find their way into the committee's ultimate report, and influence the Mayor and Transport for London. But I will mention the opinions of one Richmond councillor, their "Cycling Champion", no less, Conservative Katherine Harborne, and I will put here, in more detail, the response that I gave to her in the debate, and this will inevitably lead me into a discussion, again, of the danger of cycling in the UK.

I broach this subject again not because I wish to dwell on bad things, but because I believe, and have believed for a long time, that realism on this subject is critical to generating the environment of honest debate that must be the necessary foundation for a real political movement to restore cycling to the major component of our transport system that it once was, and needs to become again, if we are to deal with the environmental challenges of the very near future, of which the greatest is probably global warming.

I don't need to quote what Katherine Harborne said in the debate because she issued a press release in response to the Transport committee's consultation beforehand which said very similar things. I have linked there to Richmond Cycling Campaign's website, and it is interesting to read there of the other correspondence that RCC have had with her. Anyway, her press release says:

Katharine Harborne is the Cycling Czar for the London Borough of Richmond, where she is also a councillor. 
“We have no family car and never have,” she says. “Cycling gives freedom to me and my family – and keeps me fit. My teenage son is working to get on the Olympic cycling programme. 
“But we are still some way from getting cycling accepted – and more people on their bikes. In fact research from British Cycling found that nearly a million women would like to start riding a bike, but many are put off by safety concerns. 
“Despite millions of pounds being spent making our roads from 2009-2011 less dangerous, safety fears have in fact increased by half in that period. 
“A 2011 report reveals that almost half [48%] of occasional cyclists think safety is a big issue and yet less than a quarter [23%] of regular cyclists think it is. 
“So if we want to encourage more people to cycle maybe we should stop going on about how dangerous it is because, frankly, it isn’t and we’re just putting people off. 
“Making London more cycle-friendly isn’t just about cycling superhighways or giving tipper trucks and skip lorries a lot of respect at junctions – 24% of people who commute in London could cycle, but don’t. 
“Many don’t because they would get to work sweaty and have nowhere to shower or get changed. So why not encourage employers to provide more shower facilities or encourage businesses that are better equipped to Share A Shower – allowing cyclists from other firms to use their facilities?”

Lets forget about the funny business at the end of proposing that lack of shower facilities is a reason why our children, parents and neighbours don't cycle on London's roads. Does anyone think that every Dutch or Danish person who takes a 2 km bike trip at 15 km/h in their normal clothes on their sit-up-and-beg bike has a shower at the end of it?

Let's think about the meat of this whole approach, which is, and has been, shared by many others, including many prominent voices in prominent cycling organisations:
“So if we want to encourage more people to cycle maybe we should stop going on about how dangerous it is because, frankly, it isn’t and we’re just putting people off. 
Other politicians who have told us in recent times that "Cycling is not dangerous" are transport ministers Norman Baker and Mike Penning, as I reported in a previous post on a session of the Commons Transport Select Committee, at which they grievously misused statistics to try to suggest, incredibly, that the UK is doing a good job on cycle safety, better even than the Dutch. The link just quoted, to Jim Gleeson's statistics blog (Drawing) Rings Around The World, gives the truth of the matter, that cycling in the Netherlands is more than twice as safe per km travelled than cycling in the UK as measured from the official KSI statistics (Killed and Seriously Injured) issued by the two governments. But there is far more to it than this, as Jim Gleeson's subsequent post showed.

For cycling is so widespread across all age-groups in the Netherlands that a very significant number of the KSIs in that country categorised as cycling casualties are one-person incidents: basically, as I mentioned once before on this blog, Dutch cycling conditions are so safe that an awful lot of elderly Dutch people just drop dead on their bikes, or have crashes causing injury that involve no other vehicle, and these are in the statistics. We really need to look, as Jim Gleeson does, at the rates, per km travelled, of death and serious injury of cyclists who are in collision with a motor vehicle to properly assess the safety records of the two countries, in regards how it should influence decision-making on street engineering and the segregation of bikes and cars.

When we do this, the results are astounding. There are 508 KSIs involving motor vehicles per billion km cycled in the UK, compared with 61 in the Netherlands, a ratio of 8.3.

Cycling, in regards interactions with motor vehicles, is really 8.3 times safer in the Netherlands.

Their record is not just better, it is fantastically better. And fantastically better with a far wider demographic cycling, including large numbers of old, frail people who are far more likely to be killed or injured as a result of a given impact with a motor vehicle, or the ground, than the young, fit people who make up most of the British cycling demographic. And, one also might note, these are just the people, along with children, who could not be expected to mitigate danger in a purely vehicular cycling environment (that is, space shared with fast motor traffic) by cycling fast and assertively in Cyclecraft-recommended manner. So the Dutch cycling safety space is clearly a different world to that of the UK.

Apart from comparing cycle safety between different countries, which represents a contrast that will not actually be felt by British people unless they actually try to cycle in those countries and here as well, we can also compare safety of different transport modes within the UK, describing a contrast that they can feel, and react to, on a daily basis, in making their decisions on transport mode.

In the Department for Transport's Annual Report on road casualties for Great Britain 2009 (the last year for which full details are available) we discover (Table 1g) that the rate of cycle casualties (KSI) in that year was 880 per billion vehicle miles. In Table 1i we find that the rate of car user casualties (KSI) was 45 per billion vehicle miles. 880/45=19.55. Note that these are figures per billion vehicle miles, not person miles. Because average car occupancy is greater than 1, the rate per occupant is lowered, while for bikes the vehicle mile figure is the person mile figure. UK average car occupancy is 1.6. Therefore the danger ratio per person mile, bikes to cars, was in 2009 about 31.

It's about 30 times more dangerous per mile to cycle than to travel by car in the UK.

Now it is often held that the casualty rate per mile is not that important in comparing transport modes, because different modes have different uses. We expect, generally, to make longer journeys, covering therefore more miles, at least for the average person, by car or train or plane than by bike. Yet consider this. One of the main arguments you constantly hear advanced in cycle promotion is something along these lines:
Half of all car journeys in Outer London are less than two miles in length, equating to a journey of about 10 minutes by bike. 
(Cycling Revolution London, TfL, 2010)
So there's the suggestion that people in general should be perfectly happy and willing just to swap the car for the bike for all these many short journeys they do. That's telling people not to live their lives in a different way, travel less or do anything else differently, that's just telling them to go on doing exactly all the things that they are doing now, but to use a mode of transport that is 30 times more dangerous than the one they are using! Telling them just increase their risk of death or serious injury on the road, by their own choice, their own actions, by a not insignificant factor of 30!

I don't think that makes sense, frankly. I think that's treating people as idiots, as I said in the London Assembly debate. People are just not going to respond to this message. They are not going to cycle under current conditions unless they really want to cycle, or they really have to (e.g. for financial reasons), and if they really want to cycle, because they love cycling, and thus can put up with the risk, they will be doing it already. But if people with some power like Cllr Harborne stop telling people cycling is "safe", and instead do more to make it safer, then more people will give it a go, even though it will never, judging from the Dutch experience, be as safe as travelling in a car. (And the Dutch have really tried, big-time, to make cycling "safe", reducing the cycling to driving casualty ratio per km to 3.5.) Cllr Harborne, bear in mind, is she who was dismissive of the need for wider 20 mph zones in Richmond in a letter to Richmond Cycling Campaign least year.

If we are to ever make progress on this, I think the advocates of cycling need to stop treating the public like idiots who cannot correctly judge what is or is not an unacceptably dangerous activity for them to engage in. I think they can judge. I think those for whom cycling would be massively dangerous under current conditions generally make a correct judgement not to do it, and even for those who judge that it is not unacceptably dangerous, it's still far more dangerous than doing the same journey by car, and their doing it at all depends on their sense of enjoyment, judgement of convenience, and sometimes, a financial incentive. Only motorcycling is significantly more dangerous per mile than cycling (about 4 times). The danger of walking is comparable to that of cycling (Annual Report 2009 Chart 1e), but we expect journeys by foot to be shorter, and everybody walks, not a restricted demographic.

There is another argument that floats around this area, that is, that people gain in years of life from cycling, due to avoidance of heart disease and other illnesses associated with obesity, more, on average, than they loose due to death in crashes. So it's kind of healthy to cycle, even if it's dangerous. The British Medical Association have been arguing this since 1992, and the report they produced then, authored by Mayer Hillman, now out of print, but re-quoted in their 1997 document Road Transport and Health, held that:
Even in the current hostile traffic environment, the benefits gained from regular cycling [are] likely to outweigh the loss of life through cycling accidents for the population of regular cyclists. One calculation has shown the ratio to be around 20:1.
I suspect there is a lot of uncertainty here, implied in the phrase "one calculation", as there would be a lot of factors to take into account. Would, for example, more cycling mean that people walked less? In any case, this has always struck me as a rather weak promotional argument. This is, in the first place, because it involves people thinking about remote imponderables in their lives. It involves them needing to agree to make an exchange of a very clear, tangible threat in the present, of getting knocked off a bike on the road, for a promised benefit in the far future which they may well not accept as being very real. They might argue "Well if I don't die of that, I'll probably die of something else soon after", or, more rationally, they might argue that they can, and prefer to, get exactly the same health benefit from another choice of exercise or strenuous activity, whether that be running, swimming, going to the gym, or gardening, without the risk or injury or death in traffic that comes with cycling. So I don't think trying to frighten people into taking up cycling with the threat of heart disease decades down the line gets as much traction as many would hope it might.

I think we actually can get more people cycling in three ways. We can make it:

  • Relatively safer
  • Relatively more convenient (or faster)
  • Relatively cheaper

The last of these is the only one we are really doing in the UK at the moment, and more by accident than design, with rising costs of fuel and public transport. I think this is the main factor that has been driving the modern increase in cycling in London. But if we get more people on bikes by this method, we get more deaths and injuries. As even Chris Peck of the CTC acknowledged at the Assembly debate, in the last couple of years, casualty increases have outstripped the increase in cycling, despite the CTC generally in the past having been rather fervent advocates of the "Safety in Numbers" theory. I don't think that's a great policy success. I want more people to cycle, for all the health and environmental and social benefits I think it would bring, but I don't want to go back to the 1930s situation, when we had plenty of cyclists on the streets (quite how many it is difficult to tell, as I have never seen any transport statistics for the pre-war period), and eleven were being killed every week in London, about 35 times the current death rate.

I don't think non-cyclists are much influenced by how dangerous people like me or Cllr Harborne say cycling is or isn't. They do what they feel is sensible for them. No political movement ever gained any traction by telling the vast majority of the population that their basic, gut feelings about what they should do are wrong. We shouldn't expect anything different in respect of cycling. And even if that rule were proved false in this case, and the masses responded to Cllr Harbone's message (and those of others  like her) and started cycling just because they are told they should, without other policy changes to make the roads safer for them, I don't think there would be any social or moral progress there, and it's not a thing with which I would want to be associated. I want more people to cycle because they really find it pleasurable, relaxing and safe, not because they are told they should, or are dragooned into it through lack of other affordable choice.


  1. David, great read and very thought-provoking.

    Question: you brushed over the per-hour rate, but I believe that's a critical measure. Most UK people are willing to commute for up to an hour each way, max - either by foot, bike, car or train. If you don't want to drive, you put together other modes. So I could drive for 90 minutes, or put 10 minutes either side of a 55 min train ride. To understand my relative risk I need to compare 90 mins of driving vs. 20 mins cycling. Mileage is irrelevant. If I go for an hours Sunday drive vs. an hours bike ride, which is 'safer'? Clearly mileage is a distraction, as this then depends on road quality, congestion, car type, propensity to speed, etc etc...



    1. Yes Mike, I think this is a further complication that you can regard cycling either as a method of covering distance, getting places, or as an activity, a means of occupying time. I have focussed on the utility aspect of cycling, where it is basically a method of making a journey that you definitely need to make by some means or other, and there the per mile risk seems highly relevant to me.

      But if you are comparing an hour's Sunday drive versus an hour's bike ride, then you are talking about a way of occupying leisure time, and the total risk of the activity is more relevant than the per mile risk. Whatever, you certainly seem to get far more people occupying time with cycling, doing it for leisure, when you remove the risks of motor vehicle interaction, as the success of the Sustrans traffic-free cycleways show.

  2. Quite right David.

    Existing campaigners often make statements which can be reduced to "I do it, so you can do it too".

    However, by doing this they forget why it is that they are cyclists, while the majority of the population are not. Anyone who cycles now in the UK, including Cllr Harbone, is by definition easy to attract to cycling.

    In order to grow the modal share, we need cycling to become attractive to a much wider demographic, including those people who are not so easily attracted to cycling.

    Different people have a need for a different degree of safety, and in particular, subjective safety. People who are scared to cycle now will remain scared to cycle unless conditions are made much better.

    It doesn't matter how many times existing cyclists tell non-cyclists that "the water's just fine", they're not going to swim until the shark is taken out of the swimming pool.

  3. Two points if I may:

    Firstly, as someone whose job it is to propose economically justifiable transport projects, being able to include the health benefits of cycling in the Benefit:Cost ratio would be great. I agree with you that as way to encourage people to choose cycling it will never work.

    Secondly, I often experienced a form of the 'safety in numbers' effect when crossing the road with lots of other pedestrians at special events, most commonly at football matches here in Australia, where by sheer mass the crowd would stop the cars from moving. So perhaps this is what proponents of this form of 'safety' are hoping for. But you only have to have seen the footage of the car driver with murderous intent driving through the Brazillian Critical Mass ride to see how unsafe this is. Whenever I hear 'safety in numbers' I always feel like I'm being asked to be a penguin in the wild at feeding time, and cycle with the expectation that one day the leopard seal waiting just below the surface of the ice (i.e. The errant motorist) will catch me out of the hundreds of us that all jump in the water at once. I'd like to think my human life deserves more than this.

    Great post by the way. More really useful arguments for me to use to further convince my non-cycling colleagues to support proper safe cycling infrastructure.

  4. The 'safety in numbers' sounds a bit like being invited to a suicide mission; "Only 30 % of you will you will survive, but you will all be heroes.", which makes riding a bike a hard sell. I know I wouldn't have send my son out on his bike if those would have been the odds. Katharine Harborne's remarks are fine and dandy, but cycling in London is nothing like cycling in Rotterdam, where I happen to live; even driving a car is nothing like it. I, like most people, don't wan't to make myself belief I'm perfectly safe, it has to be a verifiable fact. Millions of pounds wasted on the Olympic Games; money that could have been spent on making London a liveable city, for both pedestrians and cyclists.

  5. I have a largish family, and I can't persuade any of them on a bike, even though they all have bikes that have been sitting in their sheds for years, at a time when they each may have got the urge to cycle, and gave it up after one or two rides out. Why can't I get them to break out their bikes? Because they say it's far too dangerous and no fun being terrified every time a car is heard coming up beside you.

    I cycle and I "know" it's dangerous out there.

    I only cycle because when I retired 4 years ago I bought a trike. Motorists are more wary of a pensioner on a trike as it's large and also gives the impression I am disabled. So I am relatively safer than other cyclists. Even then I have had a couple of really bad scares.

    6 weeks ago I bought a 2 wheeler. I had to learn to ride it as I have never ridden a bike before. I have made myself safe on that, because I break the law and ride on pavements. There is no way in hell anyone is going to force me to ride sharing the road with metal weapons traveling at anything up to 60mph, with what could be a lunatic in control of it.

    I am your average Joe, or Joe'ess you might say, of the general public, representing both the elderly and those that have never cycled before. I do not commute, I do not live in a major city, I just want to cycle to get my shopping, visit friends and family, and go off for fun days out. I want to take my bike on a train to other towns and know that as I get off there is going to be a safe cycling infrastructure for me to use and explore. I don't want to have to look up the safest way to get from strange A to strange B before I travel anywhere. I just want to know that a car, can't kill me.

    Now I just wonder what percentage of the population I represent. I heck of a lot I bet you.

    London and other cities might slowly get more fit, healthy, active, commuters on their bikes with the stupid Super Highways but they wont get every other Tom, Dick, or Harry, on their bikes.

    Cycling is far too dangerous for that.

  6. One-person incidents can be significant in (parts of?) UK too according to 'Annex A' of this report

  7. Absolutely right! If we want to make the roads safer, we must first face up to the fact that we need to change. International evidence shows that if you really want to make the roads safer then infrastructure is key.

  8. Indeed. I think statistical appeals to the general public are useless, because the health effects of a sedentary lifestyle usually kick in during the latter stages of a person's natural lifespan - however, those killed in road traffic collisions are often people around my age, in the prime of their lives, killed suddenly, who could have gone and lead a fulfilling life. Even though in the long term, most of my friends are more likely to die of lung cancer, cirrhosis or other lifestyle related diseases than be killed on our roads, I suspect most people are far more accepting of the inevitability of a slow death than a random and unnatural one.

    Ultimately, there is NO problem regarding cycle campaigners "dangerising cycling" putting people off, because we already have enough coverage of road deaths in the mainstream media (which will reach a much greater audience than the CTC would) to do that. I suggest we ensure we are able to portray cycling as a safe activity where the danger is imposed by motor traffic, and that we try to make highway authorities move away from "road safety" towards "health and safety" in terms of minimising danger at source.

  9. All forms of vigorous exercise have some degree of hazard. ISTR that gardening comes out worse than cycling. (Doesn't stop me cycling to my allotment however). Not that I am arguing that cycling needn't be made safer but I would still encourage non-cyclists to get a bike.

    1. Yebbut the danger of cycling does not come from the "vigorous exercise" itself (few die of a heart attack while cycling), but from the environment. Fundamental difference, isn't it? People don't mind doing quite dangerous things if it's their choice, and fundamental to the activity. So people go skydiving, rock-climbing etc. But cycling danger is mostly imposed unnecessarily on the activity by the environment it is done in, the street with all its traffic; it's a separate layer of risk, ands one that those who don't really love cycling or don't have to do it will never want to take on. Why should they?

      I find it hard to believe anyway that gardening could be more dangerous. You don't hear people say: "I would go gardening, but I'm frightened of getting hurt".

      In the end it's how people feel that counts, and while other activities are risky, there is not the wide spectrum of social benefits to, say rock-climbing, that creates the imperative to try to get people to do it, that there is with cycling. That's why we need to try much harder to get people on bikes than up mountains, or even into their gardens.

  10. Apologies if someone has already raised this, but given that "Only 6 per cent of [presumably vehicle occupant] fatalities occurred on motorways, although they took 20 per cent of traffic" (from the Department for Transport's Annual Report on road casualties for Great Britain 2009 you linked to), how does this affect your assertion that "the danger ratio per person mile, bikes to cars, was in 2009 about 31"?

    I don't see in the report where the DoT has separated out drivers' non-motorway and dual carriageway miles from all vehicle miles. Otherwise the number you've ended up with is the result of comparing oranges with apples.

    If it does include miles from motorways and other roads that for whatever reasons cyclists are unlikely to use, then perhaps the figure is skewed downward from the true figure.

    Is it instead worth comparing KSI rates per journey, rather than billion miles travelled?

    1. I agree that that affects the argument about short trips to a certain extent, as these will not be made on motorways, which are particularly safe for cars, so the non-motorway risk per mile for car occupants will be higher than I have calculated. But only a bit.

      The per trip risk is certainly several times higher for cyclists than drivers. As I pointed out in this post, in the Borough of Camden, though 3% of journeys are by bike, 19% of road casualties are cyclists, suggesting a risk per trip ratio of 6:1 bike:car.

  11. This is absolutely spot on. I don't think any amount of statistics is going to convince most people that the kind of situation cyclists routinely face are safe. I used to believe the safety in numbers story too, but I get the feeling that only works if the numbers lead to political pressure to improve provision. In other words, it's a spurious correlation. Just having lots of cyclists in the same car-centric environment doesn't make it safer. I've been cycling in London for 5 years and if anything it seems to be getting worse.