Calls for new or stricter legislation are often based on a misunderstanding of how people’s behaviour is really influenced
One of the interesting side effects of being a citizen of one country and living in another one is that it becomes possible to, as they say, compare and contrast the two. Sometimes you see large differences between the two, and there is a lot of contrast, but at other times they appear remarkably similar.
An example of the latter is the eagerness with which governments in both my native and my adoptive country resort to bringing in new laws, or to intensifying existing ones. It sounds simple: just introduce a whopping disincentive in the shape of a severe penalty, and people will keep to the straight and narrow.
The law rules
Jogchum Vrielink, a constitutional law scholar at the University of Leuven, has a reputation for drawing attention to the near-certainty with which, whenever something goes wrong in Belgium, someone’s knee jerks and they either file a complaint or indeed demand a whole new law. In a recent book, Pro Deo, he notes with a mix of bemusement and wry resignation the frequent annoyance of Belgian policy makers with the fact that the fundamental rights embodied in the constitution prevents the unbridled introduction of new statutes to address the latest form of inappropriate behaviour.
Much of this takes place in the domain of freedom of expression. In a radio show last Saturday he laid out a solid argument against new legislation to directly address phenomena like racism, sexism, holocaust denial and incitement to hate and terrorism. Really serious instances of unacceptable behaviour – intimidation, threats etc – can be dealt with effectively using existing, less specific laws, so there is no real need. But the problem is not only that it new measures are unnecessary, but also that they can backfire. For example, they risk suppressing signs of growing radicalization, thus making the work of the security services harder. They also curtail the safety valve function of racist speech, which may push some people towards more physical acts of racial abuse. Yet politicians who wish to play to the gallery are often all too keen to exploit anecdote and, in Vrielink’s words, “surf a wave of moral panic” to introduce new legislation. Unsurprisingly, Belgium possesses a remarkable assortment of illiberal laws.
By coincidence – or was it cosmic synchronicity? – the very same day a similar piece of symbolic legislation made the headlines in the UK: the penalty for drivers caught using handheld mobile phones while driving are to be doubled from £100 ($130) to £200, and from three penalty points to six. (These points stay on a driving licence for up to 11 years, and collecting 12 or more points within a period of three years lead to immediate disqualification.)
At once, critics pointed out that these new penalties were completely inadequate as a deterrent, and called for harsher treatment of offenders. Yet the UK too already has laws that can deal with dangerous driving (with penalties of up to two years imprisonment and unlimited fine, and mandatory disqualification of minimum one year) all the way up to causing death by careless or inconsiderate driving (up to five years imprisonment and a mandatory one-year disqualification) and causing death by dangerous driving (up to 14 years in gaol, an unlimited fine, and a minimum of two years disqualification).
A better message
These are not small beer, and make the action of the government somewhat puzzling. Their announcement was the start of an information campaign under the regular Think! traffic safety banner, and at first sight the doubling of the penalties for a particular offence should at least make would-be offenders sit up and take notice. But then, why not point out that the offence in question is already subject to far more significant penalties?
Actually, even that is unlikely to change many minds. Let’s assume for a moment that all prospective offenders are rational – in other words, they take into account the outcome and the likelihood of being caught. Irrespective of the actual probability, doubling the outcome should therefore have an effect on their behaviour.
In practice the fact that our information is limited matters, however: we are told the consequences of breaking the law in big fat print, but we don’t really know how likely it is that we will be caught. So we take a guess, based mostly on what we observe. We look at how many people in our circle get caught (nobody). We see what the media tell us (not much: occasionally we learn about someone being convicted, but these are rare events). At the same time we regularly hear of cuts in police numbers, so our conclusion is that the chance of being apprehended is very low indeed.
This misperception of risk, and in particular the probability component of it, in the context of committing an offence is confirmed in a study for the German Institute for the Study of Labor. Behnud Djawadi and René Fahr ran an experiment in which people had the opportunity to take bribes, but could – if caught – pay a (relatively) severe penalty. Despite this most participants happily engaged in petty corruption as if the risk was much lower than it really was.
What you don’t see is not there
A plausible explanation for this apparent risk neglect is what Daniel Kahneman called What You See Is All There Is – or, more precisely, the polar opposite formulation (‘What you don’t see is not there’). We have no evidence of what the true likelihood is of getting caught texting or tweeting behind the wheel, and we simply assume that, just like the likelihood that a meteorite will hit us, it is close enough to zero not to worry about it.
The government could seek to enforce the law more intensely. But even if the chance of being caught could be doubled (which would be unrealistically costly), this would not materially alter would-be offenders’ perception, and so the incidence of the behaviour.
The key is perception management – of the chance of being caught, not of the consequences if caught. But that would need to be done intelligently. Simply publicizing the actual conviction rates may not be the most effective way of catching the general public’s attention. Thankfully only a small percentage of all drivers commit the offence of dangerous driving, and inevitably only a fraction get convicted. That eventual ratio is unlikely to impress many people. There are over 45 million people with a driving licence in the UK, so even if a ludicrously high number of drivers (say, a million) were caught each year, that would still “only” be 2.5%.
We are much more sensitive to absolute figures that feel larger (or smaller) than what we are used to or would expect, than to relative figures. If I told you people in the UK consume 500 million litres of milk each year, would you think that is a lot, or a little? (Actually, the true figure is 5 billion litres, so it would be very little – but that is not conveyed by the huge number of 500 million).
Using this stratagem, instead of trumpeting the penalties, perhaps the government could consider a slogan like “Last month, 106 drivers were convicted because they were texting at the wheel. They were fined a total of £258,400 and received a total sentence of 131 years imprisonment. Don’t join them this month.”
Insulting a donkey
What about the Belgian law makers so keen on restricting freedom of speech? In the same radio show Jogchum Vrielink observed that there is no obvious link between antiracist legislation and public morality (perhaps on the contrary): there is no law that prohibits the use of the word ‘negro’ in the USA, yet it is rarely heard, not just in polite conversation but in any conversation. Social norms appear to be perfectly capable of performing all the behavioural influence that is required.
It should not be beyond the wit of policy makers to come up with other ways to influence public morality and behaviour than by legal repression. It is said that the law is an ass – meaning that legislation is not always sensible.
Alas, that judgement seems to apply as much, if not more, to policy makers.
Koenfucius 