Can safety regulations kill you? How safe are seatbelts and seatbelt laws?
Podcast: Download
By Navin Kumar
Article ID: 1327
Seatbelts save lives, right?
They secure people to the vehicle so that if an accident occurs, passengers are prevented from being thrown around and hitting interiors of the car and breaking their necks. They prevent passengers from crashing into each other or being thrown out of the car. Given their ability to prevent second impact, their life saving ability should be obvious.
Right?
Not really. One of the fundamental principles of economics is “People respond to incentives”. When the ‘cost’ (in terms of injury) of driving rashly is high – such as when the streets are narrow and full of potholes – people respond by driving slowly. Conversely, when the cost is lower – such as when you’re driving along a deserted, straight, smooth, well marked road – you can drive faster with less fear of killing yourself. When a car is equipped with seatbelts, people drive faster since the pain and suffering they would experience if they crashed has been reduced.
If this sounds absurd, ask yourself: would you drive with greater care if – one fine day – you woke up and discovered that someone cut away your seatbelts? If the answer is yes, then it means you aren’t driving as safely as you can, right now with your seatbelt on.
People face trade-offs. They can spend an hour cleaning up the house or watching TV, but not both. They can have, within the hour, a clean house or the pleasure of having watched Ross get divorced, but not both. Similarly, they can drive fast and get home from work quickly (risking a car crash) or they can drive slowly and reduce the risk of getting into an accident (but come home late). The speed most people finally choose depends on their tastes, urgency and whether there’s something good on TV or not. If the risk of getting hurt in an accident is reduced, people “consume” some (if not all) of the increased safety by converting it into speed.
In essence there are two ‘effects’ of having seatbelts on the number of people who die in car crashes. Seatbelts reduce the severity of accidents when they happen, and they decrease the number of deaths. But they cause people to drive faster and more recklessly, causing more accidents and increasing the number of people who die.
What then, is the net result? Do seatbelts increase or decrease the number of people who die? We can’t come to any conclusion using just theory. We need to research and at look at empirical data. And in 1976, an economist at the University of Chicago’s Graduate School of Business – Sam Peltzman – set out to do just that. He discovered that mandatory seatbelts did nothing to the death rate. In other words, the two effects cancelled each other out, resulting in neither fewer nor more deaths. More evidence came in 1982 from Professor John Adams (University College, London), who surveyed 18 countries (covering 80% of the worlds driving population). Adams concluded that countries with seat belt regulation fared no better than those without. In some cases, they fared worse. He succinctly phrased the Peltzman effect: “Protecting car occupants from the consequences of bad driving encourages bad driving.” Most studies following Peltzman have come to the same conclusion.
In the interest of fairness, there are people – like Steven Levitt- who insist that the Peltzman effect is a lot more trivial than is presented.
The Peltzman effect has many applications outside regulatory economics, like in the area of risk compensation. Researchers have discovered, for example, that wearing a helmet while bicycling can cause drivers to drive closer to you because the amount of injury you suffer as the result of a possible collision is reduced. In fields outside safety and traffic, studies have suggested that condom usage might not reduce the spread of AIDS since they encourage people to have sex more often. At more dubious levels, it has been suggested that risk compensation is why people get into more accidents near home or why so many rapists happen to be relatives or friends of the victim.
William Rodgers, the UK’s Secretary of State from 1976-1979, said this about seatbelt regulations:
“On the best available evidence of accidents in this country – evidence which has not been seriously contested – compulsion could save up to 1,000 lives and 10,000 injuries a year.”
But unfortunately, this isn’t true. His estimates are based only on the direct effect of seatbelts: their ability to reduce the impact of accidents. It doesn’t include the fact that people will change their behavior in response to changes in their environment.
What’s the moral here? The most obvious one is that we should be extremely careful about introducing new safety regulations. This covers not only traffic related regulations but also – for example – child-proof safety caps for medicines. Here’s John Stossel’s take:
“In 1972, the FDA passed a law requiring child safety caps on many medications. It was supposed to keep kids from being poisoned by drugs like aspirin. But there is an unexpected side effect. Because safety caps are hard to get off, some people – particularly older people – leave them off, and some parents, feeling safer with the cap, leave the aspirin where kids can reach it.”
A study of this “lulling effect” concluded that, because of this regulation, an additional 3,000 children have been poisoned by aspirin.
There is a far more important lesson here: whether it comes from the government or the company HR department, be skeptical of any idea or policy which changes the rules and makes a prediction which doesn’t take into account how people will behave after the rule change. The condoms-might-accelerate-the-spread-of-AIDS hypothesis is an example.
Of course, this doesn’t mean that such policies shouldn’t be put into motion. What skepticism means here is that all such policies should be monitored to verify that the Peltzman effect doesn’t defeat the policy’s purpose. The implications of ideas like Peltzman’s study aren’t that things shouldn’t be done, but rather that the ideas should be evaluated by the results they produce and that we shouldn’t be blinded by ideology – even on so “trivial” a matter as a seatbelt or a bicycle helmet.
Podcast
Support
Email alerts
RSS
Facebook