Victoria suffered just under 300 deaths in road crashes in 2010. That’s a tragedy nearly every day, but it’s still a small fraction of the toll exacted by motor vehicles 40 years ago, when the road toll peaked at 1061 in 1970 (at at time when there were fewer people and many fewer cars). I couldn’t find a graph for Victoria but here is one for Australia as a whole, showing the same pattern with a slight lag as other states followed Victoria.
Anyone my age or older will remember that, after decades of accepting steadily increasing death rates as the price of mobility, Victorian governments of both political persuasions finally took the politically courageous step of enforcing higher safety standards – first seat belts and automative design rules, then effective techniques to catch and convict speeders and drink drivers, then helmet laws and more stringent license testing, among many others. Victoria’s interventions were eventually followed by other governments in Australia and elsewhere, but the lags are such that Victoria has gone from having some of the most dangerous roads in the world to having some of the safest. Nevertheless, and not surprisingly, these steps aroused plenty of opposition at the time, and the opponents were able to produce supposed experts to back their arguments.
What might seem more surprising is that even after four decades in which their claims have been refuted beyond any reasonable doubt, the same experts are still pushing the same discredited lines, and still finding a ready audience. With a closer look at the experts and their audience, this fact is perhaps less surprising, but still requires some explanation.
The arguments against road safety interventions are of two kinds, though they often intertwine. The first involve arguments against specific interventions, for example
* Seat belts increase the risk of death because people may be trapped in their cars rather than being “thrown clear”
* Variance in speed of vehicles matters more than average speed so we shouldn’t enforce speed limits
* Speed cameras/breath test machines are unreliable and give lots of false positives
* Restrictive vehicle design rules will raise costs, leading people to buy older/cheaper cars and reducing safety
None of these arguments stand up well to scrutiny, but I don’t propose to discuss them here. I’ll set up a sandpit for people who want to argue about specific cases.
The second is a general argument, purporting to show that any regulatory intervention to increase safety will be ineffective (although it is sometimes applied inconsistently by people who oppose some interventions but not others). The central idea is that any reduction in risk below the level that would arise in the absence of intervention will lead people to take more risks, wiping out (or, in some versions, more than wiping out) the first round benefits.
This kind of argument has been advanced (apparently without much cross-acknowledgement) by economists of whom the most notable is Sam Peltzman, under the name ‘rebound effect’, and by geographers, including John Adams, under the name “risk homeostasis”. Adams in particular likes to cite “Smeed’s Law” a statistical relationship first estimated in 1949, which showed that, as the number of vehicles increased, the number of road deaths increases, but less than proportionally. Victoria fitted the Smeed’s Law pretty well until 1970, after which deaths fell sharply while the number of vehicles continued to rise. Nevertheless, Adams has continued to claim that both Smeed’s Law and risk homeostasis fit the data.
Of course, it’s not unusual to see academics pushing their pet theories long after the evidence has turned against them, and some degree of stubbornness in the face of contrary evidence is desirable – sometimes the disconfirming data is wrong, or is driven by a run of chance events. And, as anyone who has followed such debates will know, it’s always possible to tweak the data until you get the result you want. But you would think by now that the stunning success of Victoria’s interventions would have produced at least some admission that the theory and the data don’t fit too well. Not a bit of it. Adams, Peltzman and others are still behaving as if Victoria’s interventions had produced the increase in fatalities they predicted, and, as I mentioned, still getting plenty of airplay from prominent thinktanks.
The explanation of course is that Adams and Peltzman are libertarians, and the thinktanks that back them are similarly inclined. Peltzman checks just about all the US boxes – professor of economics at Chicago, fellow of AEI, Cato. Adams isn’t such a joiner, but he is clear enough on the political implications of the argument. For example, in explaining persistent belief in the effectiveness of seat belt laws, Adams writes
Why should the government be so assiduously promoting and inflating this myth? It has ready access to the numbers that disprove it. I offer a simple, cynical, explanation: it feeds the larger myth of the efficacy of government.
There are obvious reasons why libertarians would like to believe that road safety laws are ineffective and that global warming is a hoax or fraud. It is of course, possible to argue that, regardless of the benefits of seat belts, people should not be forced to wear them, but that argument doesn’t work for speed traps, RBT, and so on, unless you want to try the extreme Coasian view that such matters should be settled by voluntary agreement (Adams gives this view a nod in his paper Risky Business). Issues like road safety and global warming make it clear that our everyday actions such as driving a car impinge on each other in critical ways that can’t be resolved through the spontaneous operation of market mechanisms.
Sometimes, as with road safety, there is little alternative to direct interventions of the kind pioneered by Victoria. In other cases, as with global warming there is a choice between direct regulation (specifying permissible designs for all kinds of electrical equipment for example) and measures like carbon taxes and emissions trading which, while relying on government action in the first place, leave a lot of the hard work to market processes. A sensible libertarianism would seek to identify the latter cases and present arguments for market-oriented mechanisms.
Sadly, while there are individuals with libertarian inclinations who argue in this way, the libertarian movement as a whole has chosen the path of magical thinking, hoping that if they can keep coming up with debating points, problems like global warming will go away. The libertarian think tanks in the US and Australia are uniformly delusionist on climate change, as are the great majority of individual commentators who self-identify as libertarians .
I suspect (and hope) there may be quite a few libertarians who aren’t that comfortable with the anti-science wishful thinking displayed on these issues, but prefer not to pick a fight with their fellow libertarians on an issue that may appear peripheral to their own concerns. I would urge any such to think again. Once intellectual standards are debased in this way, the damage cannot be contained. Bad arguments are accepted because they produce comfortable conclusions, or because they are put forward by political allies. This works (in a way) as long as you can assume that all the correct answers are known, having been revealed in some sacred text or another. But they imply (and reveal in the case of climate change) a total incapacity to deal with anything new. It’s not surprising, as I mentioned not long ago, that the free-market right hasn’t come up with any new ideas in decades. Like other movements that began with a radical openness to new ideas, they have become locked into a dogmatic orthodoxy, immune from empirical refutation.
fn1. Also put forward by psychologist Gerard Wilde.
fn2. It would be similarly convenient for socialists to believe that people aren’t motivated by economic incentives (or wouldn’t be if their consciousness was properly raised) – a large part of the disaster of communism was the attempt to act on this belief.
fn3. I should say that I haven’t seen anything specific from Peltzman on climate change. But, if he believes that the thinktanks with which he is prominently associated are badly wrong on a major scientific and policy issue, he ought to say so.
fn4. I am not interested in hearing from libertarians who conform to this stereotype, but I will establish a sandpit for those who feel impelled to restate their allegiance to tribal orthodoxy (with or without hedges and qualifications). On the other hand, if anyone wants to self-identify as a libertarian who accepts mainstream science as represented by, say, the IPCC or all the scientific academies in the world, I will certainly be interested.