The case for zero tolerance

A few suggestions on driving emerged from the Safer Journeys first actions paper. While some are controversial – such as raising the minimum age to 16 from 15 – others, like better training for motorcyclists and longer supervised driving periods are good. If we think of being an expert driver is completing the 10,000 hours of doing the task – a theory that Malcolm Gladwell popularised in the book Outliers – then the more time on the road the better.

I take issue with one of the proposals though, and that’s drink-driving.

This chart below, via Offsetting Behaviour, and the Law Commission report, shows the effect of blood alcohol leel on the relative risk of fatal crash. The proposal is to both lower the limit to 0.05 (though it’s a soft proposal) and to have zero tolerance for those under 20.

The point of the chart is twofold – that older drivers are less likely to have a fatal crash, and that alcohol increases the odds of all drivers having a fatal accident. This is probably not the point the Law Commission were seeking to make – they are trying to say that older people can have a higher tolerance of alcohol than the young and still drive. That may be true – but not for tolerance reasons.

The more interesting chart for me is this one, which shows that the absolute number of people dying from alcohol infused crashes in each age range is about the same. There is a little drop off through the years, but we can put that down to other factors – such as experience, how often people are driving, when and why people are driving.

Experience: Older people will reach and exceed the 10,000 hours limit – especially professional drivers. This gives them lot more road sense and instinctive ability to survive than someone that is just starting out.

How people are driving: Being young comes with its hormones and angst – and driving is a new experience that creates a real rush. So of course younger males, in particular, are going to want to test the limits – of themselves, of their screaming friends and of their vehicle. Meanwhile those 30-50 year olds are slowed down by the  kids in the back, by a history of near misses and by a sensibility that comes with being older.

When and why people are driving: Older people are more likely to be racking up their miles by commuting – a pretty safe endeavor with low average speeds and a higher likelihood of any accident being non-fatal. Younger folk are more likely to be exploring, driving at night and driving on rural roads that they are not familiar with.

Between those three, and adding our incredibly dangerous (yet fun and beautiful) roads, it’s shocking that the chart isn’t skewed even more to the young. Perhaps it is sending a different message – perhaps we have a different problem?

Regardless – my question is why we tolerate any alcohol at all – for any age. People are dying as a result of drink driving and if the first chart is true and increasing alcohol from 0.00 increases risk, then why not stop it altogether?

It would be a lot easier to implement in practice: If you are driving then don’t drink.

It’s a lot easier for yourself, friends and establishments to police – if you drink then you are over the limit.

It’s harder to determine your BAC in the morning after, but if you are drinking that much the night before then the combination of alcohol and tiredness means that I really don’t want to share the road with you.

This is a system that already works – and works well – in heavy industry across the world. If some of the best minds on safety are mandating zero tolerance for any alcohol at, say, Rio Tinto or BHP Billiton, then why do we accept less on our roads?

One thing that many plants do is allow staff and contractors to test themselves before going on site. We can do the same with driving – either by having interlocks, or by making sure we have a cheap but reliable enough breath tester in the home.

But ultimately if we need to test ourselves before we drive, then there are greater problems to be dealt with – and a zero tolerance driving regieme will help peoole face up to those issues.

It’s not a popular cause. People like to think they can have a beer and safely drive home. But why not?

Published by Lance Wiggs


10 replies on “The case for zero tolerance”

  1. I’m all for zero.

    I am also for no driving the day after drinking. If you have a big night, you WILL be tired, you WILL not be on the top of your game and regardless of how much alcohol you have in your system, you WONT be a safe driver.

    I wonder if the stats aren’t as skewed towards youth because of exactly what you are saying. If you have experience, age and a nice car, you generally don’t view yourself as a ‘drink driver’ – and may be more likely to drink up and get in the car.

    The only thing I wonder is, is how practical zero is. Aren’t there a number of factors (like perfume etc) that can give a false reading above zero… WOuld this massively increase the amount of work for police?


  2. I’m all for stricter drink/driving laws, but I think that there should be official/approved breath-testing devices available to be purchased, especially for those times when you’re not sure if your alcohol level is back at zero. For example: if I had six beers last night, how would I be sure that I was OK to drive at 7am the next morning. Or: if I had one beer at lunch, would I be OK to drive at 6pm that evening?


  3. The best case for a zero limit is if there are lots of folks who either mistakenly drink up to 0.12 thinking they’re below 0.08 or who can’t stop themselves from getting to 0.12 once they get up to 0.05-0.08. It can’t rely on the risk reduction in the 0-0.08 range because the number of driver fatalities in that range looks indistinguishable from noise: in 2007, FOUR drivers who died in accidents were found to have BAC in the 0.03-0.08 range. If we take that as a fraction of drivers on the road in those ranges, I rather suspect we’ll have a hard time distinguishing it from the overall sober accident rate if we correct for time of day: it would be hard to tell that the alcohol contributed substantially to the accident.

    But even in the best case first noted above, we’d need to show that the risk reduction thereby achieved not only passes a cost-benefit analysis but also reduces risk more efficiently (lower cost) than other measures. The cost of the regulation is the enjoyment forgone by folks who like to have a beer or wine while out (along with enforcement costs). These benefits of a night out with a glass of wine are real and it would be presumptuous of us to say that they don’t get to count for anything. Why not? Because there are all kinds of other measures we could undertake that would reduce the risk of driving, albeit also at high cost: 10 kph speed limits, cars made of NERF, banning toddlers from being in cars without being sedated, and so on. If you say my enjoyment of a night out with a beer doesn’t get to count, I get to say that your “having” to get somewhere with your screaming toddler distracting you doesn’t get to count either. Both are completely arbitrary; the only way of avoiding that kind of arbitrariness is by letting all utils count.

    It would further seem exceedingly unlikely that overall driving risks could be reduced at lower cost by banning all alcohol consumption than by measures specifically targeted at drivers in the 0.15 and up range (where the vast bulk of fatalities happen). Think about it for a minute. What happens when police and judicial resources are wasted on very low risk drivers compared to high risk ones? We already have pretty lax enforcement on repeat drink drive offenders – lots get let out with minimal penalty for repeat drink driving on suspended driving license. Does it really seem likely that cracking down on the folks having a glass of wine with dinner does more to save lives than measures targeted at the harder core group?


    1. All good points Eric I accept and will defend that the social benefits of Alcohol are not being used in the cost-benefit analysis

      I accept that it is relatively safe for some to drive with a relatively low BAC.

      I will even accept, and this will surprise some, that it is relatively safe for some people, in some circumstances, to drive with a BAC of over 0.08. (it’s a matter of driving very very slowly in a safe environment – and even then it is still more dangerous)

      But that’s not the point. The point is any alcohol inside a driver means that they are somewhat impaired. we are simply discussing how much. It’s kind of like being somewhat pregnant.

      Major industrials will not accept anyone operating equipment or making decisions while under any influence. The owners of those plants have put considerable effort into determining the risk factors, and alcohol, at any level, comes up as a major one.

      Neither should we accept that it is ok for our children to play on the streets, for us to ride bicycles and drive cars in places where others are driving after drinking (or under any other influence). It’s hard enough to drive as it is with al the other distractions, and no matter what the stats are showing, the driver is impaired to some degree. The euphoria from that single alcoholic beverage is not helping the cause.

      Perhaps it is draconian. I agree that we don’t want Police chasing crimes that are too petty.

      So why not soften the law – make it like a speeding ticket if you are under 0.05, like a dangerous driving ticket if you are between 0.05 and 0.08 and as it is for over 0.08?

      Regardless – we should aim to make it socially unacceptable to drive after drinking. If it can work in mines based in places where drinking is about the only activity outside of work then it can certainly work in our society.


  4. @Russell – that’s interesting to know, but my point was really that there may be several examples of needing to test your alcohol level to see if it’s at zero. Based on your response, it may be possible to have a glass of wine at dinner, then go watch a movie or play, and have your alcohol level back at zero by the time to drive home.


  5. By having a limit above zero, we are sending a message that it is ok to drive after drinking some alcohol. In fact, this could be taken as the norm or expected behaviour.

    If we set the limited to zero, it is clear that it is not acceptable to drink and drive, period.

    IMHO, we will not change attitudes to drinking and driving until the law is changed to signal that driving after drinking any alcohol is not acceptable.


    1. I agree Bill – that’s exactly the point. I accept that the economic approach is to draw a line above 0.00, but the overall message is simple – don’t drink and then do things that can kill people if you get it wrong.


  6. Given the near-infinite number of potential distractions while driving, many of which often cause accidents (fiddling with the radio, reaching for a dropped CD, attending to screaming toddler), I’m not sure I understand why we need a particular bright-line rule about just one of them. Surely we need lots of simple overall messages:
    – Don’t oggle attractive pedestrians while driving
    – Don’t adjust the radio at all while driving
    – Pull over to hand the toddler his dropped toy
    – Don’t look at the GPS unit while driving
    – Don’t drive unless you’ve had 8 hours of sleep no more than 16 hours ago…
    just to name a few I noticed while driving home tonight (all blame on other drivers; I of course should not have been paying as much attention to their inattention). All of these things reduce the attention you’re paying to the road and could kill people if you get it wrong.


  7. Lance, a zero limit for all drivers is simply not go to fly with the population right now. However, I think this change does a good job of introducing us to the idea. If young drivers start driving at 16 and with a zero BAC limit until aged 20, that’s four years of zero limit driving and socialisation experience that the kids have – of going to parties and learning to say, “sorry, I’m on the water, I’m driving.” As this becomes the new norm, we might see the change you’re looking for, and that all evidence supports.


Comments are closed.