According to the theory of risk compensation, people adjust their behavior in response to perceived levels of risk; they are careful when they sense heightened risk and less careful when they feel more protected. As a result, attempts to increase safety can have the opposite effect.
For example, the much-cited paper “The Effects of Automobile Safety Regulation” claimed adding safety equipment to automobiles didn’t have the expected impact of reducing total deaths in car accidents. While fewer people in cars died, more pedestrians did. The reason was drivers felt safer and therefore took more risks and endangered more pedestrians.
A study of German taxi drivers with and without antilock brakes (ABS) came to a similar conclusion. Over the three-year study, the cab drivers with ABS accelerated faster, stopped more suddenly, and had more crashes. The study concluded those drivers took more risks in the assumption that ABS would protect them.
Economist W. Kip Viscusi coined the term Lulling Effect to describe this false sense of safety which encourages more risk-taking. He showed childproof lids on medicine bottles did not reduce poisoning rates as much as expected because parents became more complacent with the bottles without protective lids. The Lulling Effect also explains what happens when our defenses against natural disasters improve. People move into riskier areas such that deaths from floods and hurricanes have remained constant or even increased.
There are many more examples of risk compensation. Kids who wear protective gear (like helmets and knee pads) take more physical risks, people who order a healthy main dish are more likely to consume more calories (the health halo effect), and people take larger financial risks after they first start saving money.
So, what’s the best way to avoid risk compensation and increase safety?
Rather than try to legislate or punish the behavior we are trying to avoid, the most effective technique seems to be to incent for the behavior we want. Trying to improve the safety of cars for drivers or increasing the number of traffic laws may not make roads safer for everyone. However, lowering insurance premiums for safe drivers has the intended effect because it encourages drivers to make positive changes to their behavior.
In addition, people are less likely to engage in risk compensation when they feel less safe. For example, serious injuries in car crashes are rarer when the roads are icy because drivers understand the risks and drive more cautiously. We can use this approach in other situations by better educating people about potential risks. If they feel less safe, they will take less risks.
Risk compensation shows us the things we do to reduce risk may not always make us safer. It could even have the opposite result, as in the Cobra and Streisand effects. It’s laudable to try to reduce risk but beware of unintended consequences.