‘Safety in Numbers’ dismissed as a strategy by London Assembly report

I’ve written extensively about the ‘Safety in Numbers’ effect before, particularly about the folly of employing it as a safety strategy, so I won’t spend too much time going over old ground.

One of my main concerns about the ‘Safety in Numbers’ effect is that it can be stripped away from its original context – an observation of a correlation between safety and numbers, without any established mechanism for that correlation – and directly employed as a safety strategy, even as a substitute for hard action to make cycling safer. Instead of changing the physical environment of our streets, to minimise the number of interactions between motor vehicles and bicycles, and to lower the speeds when those interactions do take place, you might see politicians arguing that all we need to do is increase the numbers of people cycling. As I wrote back in October

‘Safety in numbers’ is attractive as a theory because it requires very little financial or intellectual investment in trying to improve conditions for the most vulnerable users. All that needs to be done is to ‘encourage’ more walking and cycling, without the hard effort required of improving the environment itself (which, incidentally, does have a proven record of making walking and cycling safer).

It’s also a useful way of dismissing or silencing the concerns of those who argue that cycling is too dangerous, or feels too dangerous. By talking about the hazards of cycling, and how unpleasant cycling feels, so the argument runs, we are discouraging potential cyclists, and so reducing the ‘safety in numbers’ effect, and making cycling less safe. This is something Andrew Gilligan has argued, in response to the Times’ Cities Fit for Cycling Campaign –

It’s still not clear whether The Times’ coverage will bring about many, if any, of the improvements it seeks. What it certainly will do, however, is make several hundred thousand Times readers think twice before they get on a bicycle. And if fewer people cycle, or take up cycling, the casualty rates will suffer.

That is, cycling safety is dependent on the numbers of people cycling – so stop talking about how dangerous it is.

Boris Johnson and Transport for London have placed a good deal of emphasis on ‘Safety in Numbers’ as a safety strategy. A TfL document [pdf] claims that the Boris Bike scheme will, in and of itself,

 Improve safety by increasing the number of cyclists on London’s roads

Likewise Transport for London’s 2010 Cycle Safety Action Plan [pdf] argues that

One of the most effective strategies to increase the safety of cycling may be to encourage more cycling and more cyclists.

At few years before this was written, there may have been good grounds for believing this. The rate of cycling casualties on London’s roads had generally fallen since 2000-1, over a period in which cycling increased substantially (albeit from a very low base). Yet a graph contained in the London Assembly’s newly-published report, Gearing up – An investigation into safer cycling in London [pdf] throws this relationship into serious doubt.

The casualty rate in London has risen quite consistently since 2007, despite a continuing increase in the amount of cycling over this period. Here are the figures for the average number of trips since 2000, from TfL’s Travel in London report –

And in graph form –

So the increase in cycling has been fairly continuous over the last decade, and yet since 2007 there has been a reversal in the casualty rate. More people cycling is not leading to an increase in relative safety; quite the reverse. Cycling in London is getting less safe, despite the increasing numbers.

The London Assembly report is quite right, therefore, to note that

The Mayor believes the ‘safety in numbers’ effect will improve cycling safety in London but this is not currently evident… Our analysis shows that the safety in numbers effect has not prevented an increase in the cycling casualty rate between 2007 and 2010. Therefore, there remains an imperative for the Mayor and TfL to make improving the safety of cyclists on the roads the top priority in all their cycling programmes.

Danny over at Cyclists in the City has provided an excellent summary of the main findings of the whole report, although I’d recommending reading it in full if you have the time. It’s a rallying call for action, particularly for greater financial investment in physical infrastructure for bicycles, and for the Mayor to fully commit to following London Cycling’s Go Dutch demands. The overwhelming impression it gives is that we simply cannot go on as we are, with our roads and streets so poorly designed for bicycles, and so far behind our European neighbours and other major world cities. There can be no more complacency.

This entry was posted in Boris Johnson, Go Dutch, Infrastructure, LCC, London, Safety, Safety In Numbers, Subjective safety, The Netherlands, The Times' Cities Safe for Cycling campaign, Transport for London. Bookmark the permalink.

6 Responses to ‘Safety in Numbers’ dismissed as a strategy by London Assembly report

  1. I have argued against you on this twice before, and getting SiN wrong is the main failure of the Assembly report.

    The problem is that they use Slight Injuries – notoriously unreliable – rather than KSIs, which are more reliable. (Another point is that the cycle traffic index is inadeqaute – there has been a lamentable failure to count how many cyclists are out there). Using KSI shows an overall significant decline in cyclist KSIs since 2000. Also, a trend over only 3 years does not neccessarily mean anything

    Asi have argued before, accepting SiN does NOT mean theat you don’t press ahead with reducing danger to cyclists (and others). But it does mean that you shouldn’t deny the way that road users behaviour can and does change, sometimes for the good.

    In fact iIwould have liked to see the Assembly report go a lot further in terms of its reccomendations to reduce danger from motorised traffic to cyclists and others.

    Anyway, the main thrust of the report in (i) highlighting the TfL and Boroughs failure to increase cycling’s modal share enough, (ii) caling for taking roadspace away from motorised traffic where neccessary, (iii)and calling forgetting serious money in is more than welcome – maybe we can concentrate on how we progress these items – and don’t have it as another report on the shelf – rather than going on about SiN?

    • I rather agree with your last sentence – I think we should stop talking entirely about ‘Safety in Numbers’ as a means of achieving safety, and instead concentrate on concrete strategies that will improve conditions in London, and elsewhere across Britain.

  2. Mark says:

    Of course, actual cycle KSI rates when applied to locations can be small and changes not statistically significant and so (especially in outer London) makes a pure casualty argument nigh on impossible. Coupled with the Mayor dismantling Ken Livingstones 23 funding streams (one being cycling) in favour of fewer, larger schemes, we seem to get many ‘better streets’ schemes of widened footways coated I nice paving. There is a place for these of course, but policy needs to change to force boroughs to deal with cycling infrastructure. My borough is a biking borough, but this seems to be aimed at publicity, events and a bit of parking – the status hasn’t even led to a detailed action plan.

  3. PaulM says:

    Far be it from me to challenge a road safety expert such as Dr Davis about the statistical meaningfulness (sorry about the confected word) of “slight” injuries – I am assuming that his argument is about its use in a statistical sense. However, when you are on the receiving end of a slight injury as a cyclist, I should imagine that it has a distinct disincentive effect on you, to encourage you to stop exposing yourself to further hazard. As the saying goes “it’s like banging your head against a brick wall – it’s nice when you stop”.

    Once you have been slightly injured, and thus, I suppose by definition, walked away, surely you would be most likely to conclude either that it has happened once, so it could happen again, or that you have suffered a slight injury, but next time could be more serious. Human nature being what it is, I suspect only a minority would conclude that statistically, the risk of casualty on a bicycle is quite small, only once every few squillion cycle miles, so that’s me done, I’m safe from now.

    Funnily enough, that minority includes me. My first incident involved a charming but highly confused and subsequently mortified and apologetic lady in a Nissan MIcra jumping a red light at Hyde Park Corner, bouncing me over her bonnet. I suffered a tear to my suit trousers and a badly bent Brompton but otherwise no harm done. My next two involved agressive cabbies left hooking me, with limited damage, and my last put me in hospital and through moths of physiotherapy – and yet I am still cycling.

    BUT I have to be in a minority, don’t I? And what about my friednds and family who, despite my sanguinity about the risk, take my experience as a reason to stay firmly off the sadlle?

    • Going off on a tangent, I’ve heard the same thing about the risk of casualty being once per squillion miles, or you could ride for five lifetimes without being KSI’d, or something.

      But I wonder what the ratio is for “experienced a scary event which is likely to discourage cycling” or “have to strain to go fast enough to handle stressful traffic situation”. I suspect it depends where you are, but I’d say in London it is probably nearer to once every three miles for the first one, and once every few hundred metres for the second one. (I’m sure every cyclist in the country has a long history of close passes, for example.)

      And perhaps that’s the bigger problem – the statistically low chance of getting into a serious collision isn’t really irrelevant, because it’s massively outweighed by the relatively high occurrence of thoroughly unpleasant moments, and that must discourage people from cycling more often, or indeed, ever again. (Not you though, eh!)

      Like the brick wall, most are happy to stop banging their head against it!

  4. Paul M:

    I am of course talking about this in a statistical sense. The point is that it “Slight Injuries” just doesn’t tell you whether your chances of having been hit have gone up or down. KSI does give a half way reasonable indicator, providing you have an index of exposure (time, distance, journey cycled)

    It is nothing to do with the importance of having been hit.

    As far as I’m concerned, someone overtaking too close without hitting you is a very bad thing, and does not show up on any official statistics. Also, a recorded injury (Slight or Serious) involving someone, for example, being drunk and falling off a bicycle after hitting the kerb s a record of sormthing qualititatively different from being hit by a motor vehicle being used inappropriately.

    I’m now leaving this for the reason given. . I have done a piece on the good bits in “Gearing up” and obstacles to them being acheived on http://www.rdrf.org.uk. Another piece follows.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.