Those baffling and misleading comments on Dutch cycle safety from Penning and Baker, in full

From today’s House of Commons Transport Committee on Cycling Safety. You can listen to this guff yourself, if you wish, at about the 11:56 mark.

Ellman [chair]– Can we learn anything on safe cycling from countries such as the Netherlands and Denmark?

Baker – Well, I’m always happy to look for lessons from elsewhere. I think we should be always open to that. And I’ve been over to look at cycling in Holland, in particular, which is very well known for that. I think that my colleague Mike referred earlier on to the rate per 100,000 of the population, in terms of cycle deaths, and we actually come above the Netherlands. We’ve got a better record on that. So what we can learn from the Netherlands, in my view, is probably not safety issues, particularly; what we can learn from the Netherlands is how to encourage people to cycle more, to improve the infrastructure, the public infrastructure, the public realm, to join up different modes of transport, like rail and cycle. That’s what we can learn from the Netherlands, rather than safety. I mean, I went to, I think it was Leiden station, and I think I’m right in saying that when I got to Leiden station – a medium-sized town – there’s something like 13,000 bicycles parked there every day. And no cars. Or hardly any cars. We’re never going to get to that situation, but we can make a lot more progress on that. So they’re the lessons we can learn, I think, rather than necessarily safety lessons.

Why stop there? If nobody cycled in Britain, we’d be infinitely safer for cycling than the Netherlands. A new policy strategy clearly beckons.

And then Penning’s response –

Penning – I think there’s a classic example, where as you massively increase the amount of people that cycle, your figures for deaths… [trails off, consults paperwork] For instance, on the European table I have here, the Netherlands is fourth from the bottom, with 0.84 [deaths] per 100,000 population. Where we are, I think, is seventh, with 0.71 [deaths per 100,000 population]. That is not because they don’t care about cycle safety, that is because there are so many people cycling in the Netherlands. So you will get those ratios going up. I think the Netherlands may want to come and see us, to see how we are making sure that so few people are killed cycling, in terms… as we increase the numbers of people cycling, because the figures would indicate we’re doing a bit better than they are [smug look, and smile]

So cycling gets safer the more people do it, except it doesn’t, because more people get killed in the Netherlands, because more people cycle there, except that we’re increasing our numbers here, so we’re safer. So we should stop people cycling. I think.

Or something.

I was going to attempt to pick the bones out of this, but as I’ve typed it up, I’ve come to the conclusion that it’s self-contradictory gibberish that doesn’t really merit any response whatsoever.

It’s frankly insulting that transport ministers can turn up to a meeting about cycling safety and spout this kind of evasive, dishonest and misleading rubbish. They just don’t care.


As Jim points out in the comments below, per km travelled (a sensible measure, rather than the meaningless ‘absolute number’ measure Penning and Baker decided to use) Dutch cyclists are more than twice as safe as British ones, with 9 fatalities per billion km cycled in 2009, compared with 21 fatalities per billion km cycled over the same period in the UK. His full analysis is here.

This entry was posted in Cycling policy, Infrastructure, Road safety, Targets, The Netherlands, The Times' Cities Safe for Cycling campaign. Bookmark the permalink.

38 Responses to Those baffling and misleading comments on Dutch cycle safety from Penning and Baker, in full

  1. Hi Mark,

    I am slightly confused. Are they saying that for every 100,000 people who cycle 0.71(UK) 0.84(NL) die? Or is it population as a whole? Am I being dense?

    • Hi Jono,

      They’ve taken the population as a whole, and the the number of cycle deaths as a whole, and worked out the number of deaths per 100,000 head of population, for each country. That’s how you arrive at such a misleading statistic, given that so few people cycle here, and so many more people cycle in the Netherlands.

  2. Joe Grey says:

    Transport fatalities are normally done as a factor of total kilometres travelled. So x fatalities per billion km or something like that. Baker’s response is idiotic and I’d expect more from him – or his civil servants who probably wrote it.

  3. Jim says:

    Fatality rates for cyclists per kilometre travelled are more than twice as high in Britain as in the Netherlands:

  4. Paul M says:

    Its per head of population. In the Nl, that might almost translate to per head of cycling population, but that doesn’t apply here!

    It reminds me of a very old joke. Two men are stting in a train compartment, travelling through Watford. One is tearing up a copy of “The Times” into small pieces and throwing them out of the window.

    “What ARE you doing?” asks the other. His reply: “It’s to ward off elephants”.

    “But there ARE no elephants in Watford”

    “Effective, isn’t it?

  5. I agree with Mark, having read that on the way back to the office I’m baffled….I wonder if David Hembrow would agree with their analysis of the figures and which country is safest?

  6. amcambike says:

    Cyclist fatalities increased last year in the Netherlands, by 23%, and serious injuries to cyclists have been rising since 2002.
    There is clearly something wrong with road safety policies in the Netherlands, so it is inappropriate to hold them up as an example. And as for the quality of Dutch politicians: the transport minister blamed the increased deaths on elderly cyclists, turning corners too fast on electric bikes.

    • Dutch Cycling fatalities fell from 185 in 2009 to 162 in 2010; they went back up to 200 in 2011. In that context the figure of 23% you quote looks much more like background fluctuation, than a dramatic increase.

      I certainly don’t think plucking a figure for one year out of the air demonstrates that there is ‘something clearly wrong’ with road safety policies in the Netherlands, especially when – as you hint at – a large proportion of that increase in cycling fatalities were single-person crashes amongst the over-65s.

      • Classic Amcambike 🙂

      • amcambike says:

        Fatalities did rise both among cyclists, and the over-65’s.
        Onder fietsers nam het aantal dodelijke verkeersslachtoffers het sterkst toe. Het aantal slachtoffers nam toe met 23 % (van 162 naar 200). Bij 65-plussers was dit 38% (van 93 naar 128). Het aantal doden onder bestuurders van gemotoriseerde invalidenvoertuigen (scootmobielen, elektrische rolstoelen en gehandicaptenvoertuigen met gesloten carrosserie) steeg van 23 in 2010 naar 31 in 2011. Van de slachtoffers was 80% 65-plusser. Onder inzittenden van personenauto’s en motorrijders zette de dalende trend van voorgaande jaren door. Het dodental onder voetgangers bleef nagenoeg gelijk (van 72 naar 74). Wel steeg het aandeel 65-plussers in die groep.

        Figures for last 5 years here:

        Click to access lp-i-m-0000001991.pdf

        The figures do indicate that there is something wrong with current policy, and that was recognised by the transport ministry and road safety organisations. There is a tendency to take the Netherlands as a shining example of all that is good in cycling policy, and that is unjustified. Baker and Pennings are in fact right, that cycle fatality rates are higher in the Netherlands, and that is indeed the price of a higher cycling rate, since cyclists are a vulnerably category of road user. There is no point in ignoring that fact.

        • Fatalities amongst the over-65s rose by 35, from 93 in 2010 to 128 in 2011, while the overall increase was from 162 cyclist deaths in 2010 to 200 in 2011 – an increase of 38. Quite clearly the vast majority of the increase in Dutch cycle fatalities is in this age group. It’s also worth noting that 64% of all Dutch cycling fatalities in 2011 were over-65s, and that – as you say yourself – single party accidents in this age group are the main reason for the increase in Dutch cycle injuries.

          ‘Baker and Penning are in fact right, that cycle fatality rates are higher in the Netherlands’.

          Do you understand the difference between ‘number’ and ‘rate’?

  7. Paul says:

    It’s insulting, you’re quite right. The British public (the whole British public, not just cyclists!) are being betrayed by attitudes and policies that demonstrably damage our urban environments, lead to deaths, air pollution, congestion and community disintegration at the hands of the motor car. Nothing we saw today indicated any change of heart on this issue by ministers or Government.

  8. Luke says:

    Ignoring for a moment the unbelievable statistical ignorance of Baker and Penning, and the slight error of Amcam, that figure about over 65s is startling, though not for the reasons Amcam thinks. Think about it – there are enough over 65s cycling in the Netherlands for accidents involving them to make a significant statistical difference. How many over 65s do you see on bikes here?

  9. Mike says:

    One other confounding factor that may (and I say may for emphasis again) affect the Dutch numbers is that if they have an ageing population, and that population continues to cycle (because it’s safe) then rates may go up because riders are becoming more frail.

    Quite how you’d account for that I don’t know, you’d probably have to account for long-term changes in the age demographic of cyclists and match it to the age groups suffering fatal accidents.

    Regardless of that, anyone who doesn’t understand the difference between raw numbers and rates really needs to take a very basic lesson in stats before they start making comments about relatively safety.

    • amcambike says:

      The rate cited by Baker and Penning is the fatality rate per 100 000 population. That is a rate, not a number. It is not the same as a rate per billion km cycled, but it is still a rate.

      • Oh, it’s a rate, alright, but it’s a completely bogus way of assessing or comparing safety, which is what Penning and Baker were attempting to do, as I’m sure you are well aware. For instance, the rate of cycling deaths on motorways is zero per 100,000 population, whereas on B-roads it is undoubtedly signficantly higher.

        Should that be a reliable measure of how safe motorways are to cycle on, compared to B-roads?

        Please desist.

    • Joe Dunckley says:

      Indeed. The reason that cycling in the UK is, by their metric, safe, is not merely that fewer people are doing it, but that it’s the most vulnerable — children and the elderly — who are disproportionately excluded from it by the conditions. What would our safety record look like if we had our current cycling rates but a more balanced demographic? You can perhaps extend your road safety policy recommendation for the Dutch from merely discouraging cycling to especially strongly discouraging cycling for children and elderly.

      I must at least congratulate Penning and Baker for coming up with a statement so utterly bonkers that I couldn’t possibly have predicted it.

  10. amcambike says:

    The source for my quote on rising cyclist injuries since 2002 is the transport ministers 2011 letter to parliament on road safety, as linked in my blog post. It is also the source for the details on age groups:

    Click to access verzamelbrief-verkeersveiligheid.pdf

    Terwijl bij de andere vervoerwijzen sprake is van een daling of stabilisering, is sinds 2002 bij de fietsers een duidelijk opwaartse trend te zien. De belangrijkste oorzaak hiervan is de sterke toename van het aantal enkelvoudige ongevallen, in het bijzonder met fietsers van 50 jaar en ouder. De oververtegenwoordiging van deze leeftijdscategorie hangt samen met de vergrijzing. Bij valpartijen met de fiets is het letselrisico van ouderen hoger vanwege fysieke beperkingen. In 2009 was het aantal ernstig gewonden bij ongevallen zonder betrokkenheid van een motorvoertuig (9.400) voor het eerst groter dan bij het totaal aantal ongevallen met betrokkenheid van een motorvoertuig (9.100).

  11. Luke says:

    Thinking about this, can this really be an innocent mistake? I know we should all look at evidence and not just rely on first impressions, but imagine you’re a minister preparing for debate about cycling. Your civil servant who’s been told to look up some facts and figures tells you: “Cycling is more dangerous in the Netherlands.” Surely your reaction would be “Carruthers, are you sure? Don’t they have separate lanes ‘n’ stuff?” You’d expect a bit of head scratching, and then someone says “Hang on, more people ride bikes there…” Carruthers is sent away to do more research, and the minister doesn’t say something of unbelievable stupidity.

    And if it is an innocent mistake, that’s almost worse.

  12. Norm says:

    On top of the use of the wrong ratio, isn’t Penning comparing apples and oranges here? Given the vast differences in infrastructure, isn’t there a much larger proportion of fatalities due to motorist-cyclist collisions in the UK? Anyone got those figures?

    • The best I can do at the moment, by using the SWOV data, is that of the 138 cycling fatalities in 2009, only 40, or 29%, involved ‘motorised traffic’.

    • amcambike says:

      Diagram 2 on this SWOV paper has the Netherlands data, not just for one year, but from 1983 to 2005. It is based on hospital admissions, and gives a more complete picture of injuries. The trend is very clear, and dates back to the 1980’s. Most cyclist injuries result from accidents with no motor vehicle involved. That also illustrates the limits of segregation as a safety measure.

      • ‘Most cyclist injuries result from accidents with no motor vehicle involved. That also illustrates the limits of segregation as a safety measure.’

        If a man had an accident while cycling in a field, would that illustrate the limits of segregation as a safety measure?

        Human beings are flawed. They will make errors. To the extent that people are injuring themselves in the absence of any other vehicle illustrates the limits of human beings, not the limits of segregation.

        If people are this accident-prone, how much higher would the accident rate be if they were having their wobbles and misjudgements in the midst of motor vehicles?

      • Luke says:

        Amcam, as and when most bike accidents in britain do not involve motor vehicles, I’d be pleased. When most fatalities did not involve MVs, I’d be ecstatic. Falling off my bike or getting bumped by another cyclist or pedestrian is embarrassing or annoying but rarely fatal. I’m happy to risk a broken collar bone arm (I’ve had both, not cycling). It’s dying under an HGV that worries me (and I don’t care whose fault it is).

  13. Edward says:

    Anybody who has even seen a picture of a Dutch street or sample of cycling infrastructure would surely realise intuitively that what Penning and Baker said was not just wrong but silly. It is embarrassing and spreading very quickly around the blogosphere. What is tragic is that it will inevitably lead to more pointless and time-wasting debates (rather like the one above involving amcambike) about rates and numbers and statistics instead of real progress being made.

  14. If that is your Road Safety minister then he should be sacked. He is obviously a dud with figures, cares little to check things out before saying them, has vetoed action on important safety measures and won’t act to dramatically increase the woeful underfunding of cycle infrastructure.

    As far as I’m concerned the British cycling community should focus all their efforts on just one thing until it is achieved: get Penning removed!

  15. fonant says:

    Baker and Penning aren’t completely stupid, they have simply re-stated their policy for cycling: minimise the number of people cycling at all costs, so that the number of cyclist deaths and injuries are also minimised. This is a logically valid policy, and would certainly work if your aims are limited to solely reducing the absolute number of cyclists killed and injured on our roads. Where they are completely stupid is in thinking that this is a desirable aim.

    The same policy applies to pedestrians, too. Government wants everyone to travel wrapped in their own one-tonne protective box, complete with seatbelts and air bags. Ideally in a manner that maximises profits for the governments business friends.

    Unfortunately for the government, cycling and walking are nice cheap sustainable healthy space-efficient transport options, and as such they can only become more popular as motoring becomes more expensive and congested.

  16. amcambike says:

    Pennings last point in the quotes above is absurd, because UK policy is not ‘increasing cycling’, so he can not show how he is “making sure that so few people are killed cycling, in terms… as we increase the numbers of people cycling”. His assumption that the UK can do that – i.e. increase cycling without a proportional increase in cyclist fatalities – is also unfounded. It implies that the UK has some magic formula, and he would need to say what that is.

    Otherwise, however, what the two politicians said about safety is correct, allowing for imprecise statistics. Pennings correctly recognizes that the higher cyclist fatality rate (per 100 000 inhabitants) in the Netherlands, results from more people cycling in that country. He correctly recognises that cyclist fatalities would tend to rise, if more people cycled in the UK.

    Many younger cycling advocates in the UK, treat the Netherlands as a semi-sacred example. That makes it difficult to approach cross-country comparisons rationally. After years of decline road deaths rose last year in the Netherlands, Belgium, and Germany, and that is cause for concern. The Dutch statistics also show that the long-term decline in injuries reversed earlier, so it probably is not a one-off result. Governments should look at the causes for this apparent reversal of the downward trend, and consider innovative safety policies, if the old ones are no longer effective. It is not rational to simply treat Dutch policies as a sacred cow, and to take offence if anyone dares to question their effectiveness.

    Good cross-country comparisons need to look at the full picture, and to consider possible negative feedback. The high fatality and injury risks among elderly cyclists in the Netherlands is an example. In contrast to the UK, they were encouraged to cycle, and the Dutch recreational cycle network is largely aimed at families and older people. Probably, the health benefits of more exercise among older cyclists outweigh the injury risk, but that needs research to confirm it. It is certainly irrational to wish that the UK had the same fatality / injury pattern as the Netherlands, as some comments here seem to imply. The desirable cyclist fatality rate for the UK is not the Netherlands rate – it is zero.

    • fonant says:

      Oh dear, I’m replying to Amcambike, who genuinely seems to believe that the Netherlands is a country where it’s bad to ride a bike. I would love Amcambike to give an example of a country where cycling is thriving and safe, that can replace the Netherlands as our cycling policy role model?

      I personally advocate following the example of the Netherlands for one very simple reason: cycling is an ordinary mode of transport in that country, with all the benefits that brings to individuals and the population as a whole. Of course they have more people killed and injured while cycling, just as a country that had a high percentage of the population playing chess would see higher rates of deaths to chess players.

      I’m flattered that you might think it’ “young” – in fact I used to be a vehicular cycling advocate, until I grew up and gained a family who would like to cycle but can’t because of the fear of death or serious injury from motor vehicles 🙂 I’ve been a “cycle campaigner” for many decades, for what it’s worth, which sadly isn’t much as the general population are not convinced by statistical arguments that “cycling is safe” when it clearly isn’t.

      As Amcambike almost certainly knows, research carried out by the BMA in “Cycling towards Health and Safety”, the health benefits of cycling do indeed significantly outweigh the injury risk, even in UK cycling conditions where even expert cyclists get killed by motorists. In Dutch conditions, where Safety in NUmbers might actually have a change of affecting things, and where fast motor vehicles and cyclists are carefully kept apart, the health benefits of cycling can only be a bigger multiple of the risks.

    • Otherwise, however, what the two politicians said about safety is correct, allowing for imprecise statistics. Pennings correctly recognizes that the higher cyclist fatality rate (per 100 000 inhabitants) in the Netherlands, results from more people cycling in that country. He correctly recognises that cyclist fatalities would tend to rise, if more people cycled in the UK.

      Yes, and pedestrian fatalities would rise if many more people walked, instead of driving. Indeed, I dare say that if everyone walked and cycled, and no longer drove cars, walking and cycling would suddenly become infinitely more dangerous than driving.

      Why you continue to present this absurd form of reasoning as justifiable I don’t know, but to save time any more attempts will be deleted. I’ve already asked you to desist. Please do so.

  17. Barry Kenealy says:

    I look forward to all those people from the Netherlands coming over to see how good the UK is at making cycling safe. Do you think there is a danger they may try and steal away the talented politicians like Pennings that are responsible for this resounding success ?

    • oscar says:

      I lived in the Netherlands for 27 years before moving to the UK 22 years ago. I gave up cycling almost immediately after a few close calls, but decided to take it up again a couple of years ago. Now I commute daily app. 10 miles each way. I can probably think of at least one hazardous incident every week involving MVs. For example, a taxi turned left without indicating when I was coming up behind it last week. Only life-long cycling skills, having started cycling at a very young age, allowed me to avoid a collision that day. I cannot remember having ever experienced such incidents in Holland. As a 13 year old kid I cycled regularly to the beach over 10 miles away, often with a surfboard trailer, without any incident. I believe that was possible only because most of the route was on cycling lanes separate from other traffic (including pedestrians btw).

      As fonant said above, this argument only distracts from making Britain’s roads safer for cyclists. Who knows what Pennings’ motivation for using that statistic was. If it was to diffuse the campaign, I must say it may have worked.

  18. Jonathan says:

    The Ministers’ frank but misguided comments clearly highlight the difference between Road Safety and Road Danger Reduction – Road Safety advocates think there’s no problem if all the pedestrians and cyclists are scared off our roads, whereas Road Danger Reduction advocates want to address the problem causers, ie. the motor vehicles with all the size, weight, speed and therefore momentum.

    Here’s a lesson in statistics for Mr Baker and Mr Penning:
    The local authority where I live had no road fatalities in 2010, and this figure rose dramatically to 8 road deaths in 2011. This shows an infinite rise, and since no deaths were caused by collisions involving only pedestrians, the only way to address this shocking trend is to ban all but pedestrians from our roads.

    /walks of dreaming of a world with less dangerous roads, no vehicle noise and reduced air pollution…

  19. Wyadvd says:

    I agree that when looking at how safe an activity is, then a risk ?(normalised for the relative popularity of that activity) should be used . A suitable measure for comparison ocross borders or modes is deaths per million passenger km. I have done a few figures and when the very safe roads from which only motorists can benefit(motorways) from are taken from the figures , then cycling in the uk looks less risky per million km travelled than driving in the Netherlands! Explain that Penning you idiot !

  20. Pingback: What do they have in common? | Road Danger Reduction Forum

  21. Pingback: Norman Baker playing with statistics again | As Easy As Riding A Bike

  22. Pingback: The going rate | As Easy As Riding A Bike

  23. Pingback: How to lie with statistics. | Bike-Riding Motorist

Leave a Reply

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

You are commenting using your 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.