Cars, phones, and dirigiste policy

By Eric Crampton 28/10/2013

Here’s public health specialist Nick Wilson on a potential measure to stop drivers from using their phones.

But it is possible that there are other options to be explored, including ones that we canvas in the NZ Medical Journal piece. One of these options could be a requirement that all new cars imported into NZ (eg, from the year 2018) could be required to have technology that automatically stops mobile phones in the vehicle from ringing when the vehicle is in motion. That is “smartcars” could automatically turn off “smartphones”.

A non-exhaustive short list of reasons why this seems a bad idea:

  • Unless you’re relying on a Bluetooth connection with the car, you’d need some kind of compliant phone that automatically interfaces with the car. International phone manufacturers and car companies don’t exactly jump when NZ says so.
  • Bluetooth is sufficiently finicky that it wouldn’t work automatically.
  • If you are connected with your car via Bluetooth, it’s typically to run a hands-free calling system with your radio. Phone rings, you hear it through the radio, you say “Answer”, it routes the call through the radio’s speakers. The system that would let you shut down call answering is the one that enables hands-free answering. 
  • The first thing I would do if Nick got his way on this would be to download an app blocking it. 
  • How could the car tell which phone belongs to a driver and which belongs to a passenger? Do you want to ban passengers from talking on the phone too?
  • If NZ is the only country adopting this idea, you’re going to have a hard time convincing any car manufacturer to support the tech. Best you could then hope for is something that gets added onto the car at point of import. Something that could be ripped out of the car by me when that happens – sheer bloodymindedness can be a powerful motivation for learning which bit of electronics needs to be circumvented.
  • It’ll hike the cost of new cars compared to used ones, encouraging people to keep older cars on the road for longer, worsening our fleet average age and our emissions profile.
  • Talking is hardly the most distracting thing you can do with a cellphone while driving. Reading Twitter is another. Will the system have to decide which apps are allowed and which ones aren’t? How, when there are a billion potential apps out there?
  • Get off my lawn and get out of my car. That can be a reason. One that isn’t given enough weight these days.
I’m sure you can come up with other reasons. If something like this ever goes through, I expect it’ll be with strong support from Navman and the other SatNav companies, who’ll lobby for complete blocking of all apps, including (especially) Google Maps. 

If we really want to hit the biggest source of in-car distraction, there is really only one solution. The back seat needs to be separated from the front section. Preferably in its own separate bubble.

0 Responses to “Cars, phones, and dirigiste policy”

  • It is good to get critical comment on the mobile phones while driving issue. It is a significant problem as there were 59 deaths and injuries associated with mobile phone use in NZ cars in 2012 according to the NZTA (Dominion Post 22 October 2013). Some of these casualties are among those not using mobile phones but who are killed or injured by mobile phone users – so that is probably also a concern for many (including for libertarians).

    In considering a possible “smartcar–smartphone” solution it is hard to predict various moves and countermoves, especially given technological developments. In general I suspect that smart technology can usually achieve the desired goals eg, a review of ignition interlock devices in cars in the US indicates success in preventing repeat drunk driving (see this systematic review:, presumably despite people trying to outsmart these devices.

    Nevertheless, other non-technological solutions might ultimately prove to be more cost-effective in reducing mobile phone use while driving (eg, better enforcement and higher fines). What’s probably needed is a thoughtful societal discussion about the options and the pros and cons of each option. We should not be too quick to rule out any particular option – given the complexities and trade-offs that need to be made between convenience, cost and preventable deaths.

  • I haven’t ready access to the 2012 numbers. But in the 2011 calendar year there were 259 fatal crashes and 9545 injury crashes resulting in 284 deaths and 12,574 injuries. [Source: ] I will assume that the proportions are identical across deaths and injuries in 2012. In that case, your 59 incidents would involve 1.33 deaths and 57.66 injuries: let’s round that to 2 deaths and 56 injuries.

    NZTA breaks things down further. There were 2060 serious injuries and 10514 minor ones. Let’s assume this proportion here holds as well, rounding up in your favour. We then have 2 deaths, 11 serious injuries, and 45 minor injuries.

    And we’ll further assume that they were all entirely caused by the cell phones.

    Further, let’s assume that every one of the people killed or injured was somebody outside the driver’s car. I don’t want to get into a fight with you about whether we should count costs the driver imposes on himself or his passengers – it’s tangential to this and distracting. So I’ll assume every one of those is an utterly innocent bystander.

    NZTA reckons the social cost of a fatality at $3.8 million, a serious injury at $650k, and a minor one at $64k. So that’s $17.6 million in social costs.

    The greatest benefit your proposals then can provide is $17.6 million. Let’s round it up to $20 million. I’ve specified that all of these costs fall on us all, random-draw. Per person, then, for 4.4 million people, that’s a bit over $4 per capita. Let’s round that up to $5. Because people used cell phones in 2012, each of us bore costs on the order of $5. Certainly no more than $5 – look at all of the assumptions I made in your favour to get here.

    Now let’s think about the costs you seek to impose on us.

    Every passenger in every car and every bus would be unable to use their cell phone while commuting, for any purpose other than calling emergency services. Every driver that wants to use his phone for satellite navigation wouldn’t be able to use it because you went and banned phones from working in cars – a dedicated ban-avoiding GPS costs at least $100.

    Say that only half of the country ever uses any kind of motorised transport – the other half only walk. And say that only half of everybody in cars has a smartphone (ANZA estimates 51% have smartphones). Then only a quarter of the population is getting benefits from cell phone use in cars. If each of them gets at least $20 worth of use from their phones while in cars, you will have done harm even if your policy came at zero implementation cost.

    To save $5 per person in New Zealand – less than my daily coffee spend – you are willing to impose a pretty substantial burden on all of us.

    Nowhere in here did I get into libertarian arguments. I’m a libertarian pluralist – I’m happy with restrictions on my liberty where there’s a compelling case for it. But I’m really really certain that your “ban car or phones that Nick Wilson doesn’t like” policy would wind up costing at least 10 times more than it could possibly provide in benefits.

    It isn’t just the “how much it costs Parliament to pass a law” cost that you cite. It’s not even just the enforcement costs. Think about what it would do to car imports to impose your ban. Think about what it would do to the market in cell phones and to our parallel import abilities to impose your ban. Think about the carpoolers who are able to get some work done in the car or on the bus because they can use the internet while driving. Think about people who use their cell phones as media player after docking it in the car. Think about all of the little bits of the good life that you would be thwarting with your ban – all in pursuit of $5 per capita.

  • flight mode is a thing Eric.

    Stopping reception and transmission of text and voice services does not need to disrupt other services from the phone.

    Which is not to undermine your premise, but to throw some technical reality in there.

  • Flight mode exists, sure. I expect that Nick wouldn’t be happy with drivers using their phones, in flight mode, to play Sudoku games while stuck in traffic though. Then you need some mechanism not only to activate flight mode while driving but also to block apps. And to avoid doing the same to passengers’ phones.

  • The general approach suggested by Eric is indeed a rational one ie, weighing the valuation of convenience to drivers in permitting phone use compared to the cost of health harm from crashes. Some additional points though:

    • Previous NZ data indicated that the use of mobile phones by drivers contributed to 25 fatal crashes and 482 injuries over the period of 2003 to 2008 (Minister for Transport Safety. Land Transport ( Road User ) Amendment Rule [2009]. So if we use this ratio on the NZTA figures for 2012 (of n=59 deaths and injuries) it might suggest 3 deaths and 56 injuries per year (ie, non-trivial in my view). It is likely however, that these could be underestimates as for deaths it might sometimes be hard to establish the role of cell phone use and if people are non-fatally injured they may remove evidence of any phone usage (given it is now illegal).
    • Another factor that needs consideration is that adverse impacts on driving performance from mobile phone use may contribute to traffic congestion and hence waste commuter time (as per this study: Of course crashes from mobile phone use will also waste commuter time (especially if the crashes block motorways). These harms could probably be potentially be quantified with sophisticated modelling – and added into the equation.
    • On the inconvenience side of the ledger, I suspect this can be minimised with any smartcar-induced phone disabling being optimised in various ways. Eg, not applying to GPS, not for dialling emergency numbers etc etc. If we now have some cars that are smart enough to park themselves – then the technical issues are likely to be fixable.

    So these factors suggest to me that it is not obvious that further controls are not desirable – and that the debate would certainly benefit from more information. Also I have no default view in favour of a new smartcar-smartphone law – it is just one of various options that could be considered. Indeed, in other road safety domains I suspect there is legislative over-reach and that the bicycle helmet law is a mistake (from a cycling normalisation perspective). But overall I don’t think that mobile phone use by drivers should be trivialised and it is also part of a larger problem of driver distraction. In the observational study we conducted in Wellington there were people observed with takeaway meals and even books resting on their steering wheels. So the culture of highly distracted driving seems to be problematic – something that the current law against mobile phone use by drivers may have started to address.

    • 1) You can’t use data from the period prior to the ban on driver cell phone use to justify further bans. You have to use post-ban data.
      2) Even if we put it up to 3/56, I’d be shocked if the costs wound up higher than I’d put them. Recall I was saying phones were 100% causal in 2 cases rather than partial contributors (which they would have been).
      3) MoT runs per-crash numbers that include these congestion costs; that’s why I rounded up to $20 million. I used the per-fatality/injury instead.
      4) You need the car to be able to tell whether the phone is being used by the driver or a passenger. I expect that’s non-trivial. As in “It would cost hundreds of dollars at least per car to run that kind of system, if it is at all possible”.
      5) What’s wrong with just relying on more general rules/fines for distracted driving?

  • response to 5) – its post-hoc. You generally end up identifying “distracted” drivers after a crash event.