Unfortunately, it's difficult to disentangle the Australian experience, really, to answer your question, because one of the key things is that although when random breath testing was introduced across the states it was the initiative that caught a lot of public attention and really, I think, the driving force behind the general deterrent effect, at the same time, we enhanced our penalties. It was very much the case that mandatory loss of licence, licence disqualification, was introduced for drunk drivers.
Now effectively, in all the states, with a few anomalies, if you get caught for drink driving in Australia, you will lose your licence. We have a graduated set of penalties, so the higher your blood alcohol concentration, the higher the penalty will be. That includes the fine and also the period of disqualification. There's no doubt that the threat of losing your licence, that general deterrent threat, has had a big part in our success in terms of reducing drink driving. At a public level, that has largely been through the highly visible random breath testing, but that threat had teeth to it, had meaning, because drivers were concerned about losing their licences. I think the two go together, and I would certainly be encouraging that.
As a psychologist, I would argue that you want to ensure that there is a higher degree of unpredictability in the activity but a high degree of certainty that if you are detected you will be punished, and that punishment will be reasonably severe. In fact, the literature suggests that it's the certainty of the punishment and not necessarily the severity that's most important.
I would certainly be encouraging you to think about them as a package of initiatives in which you have the random breath testing to increase that general deterrent effect, and some form of mandatory penalty, which means that the likelihood that drivers will lose their licences if it is detected is very high and in fact there's a very high certainty of some form of penalty.