Khyaati Acharya wonders if society's drive to reduce risk actually increases risk-taking behaviour, suppressing 'consequences' from choice

By Khyaati Acharya* 

Economic theory is largely concerned with how changes in incentives alter human behaviour. Take risk compensation theory for example. Risk compensation theory suggests that humans tend to adjust their behaviour according to a perceived level of risk. This sees people engage in riskier behaviour when they feel safer while becoming more risk-averse when they perceive higher threats.

That more stringent safety regulations may actually result in increased risk of accidents is consistent with many other examples of well-intended regulations that have resulted in unintended consequences. Consequences that, occasionally, may either partially or completely offset the effects of the regulation, sometimes inflicting more harm than help.

The reason for these counterintuitive effects being the change in incentives that alters human behaviour. And this, as economists Russell S. Sobel and Todd M. Nesbit posit, is “The Holy Grail of economic research: to find a theoretically sound model that predicts a counterintuitive result with a simple incentive story”.

Examples include the minimum wage and the notion that it may just harm those that it is intended to help, environmental policies that actually hurt protected flora and fauna and mandatory bicycle helmet laws that tempt risky behaviour and lead to more accidents.

One of the most famous examples of this is the Peltzman Effect. In recognition of Sam Peltzman, a professor of economics at the University Of Chicago Booth School Of Business, it refers to the notion that drivers may engage in riskier driving in response to increased automobile safety.

‘The Effects of Automobile Safety Regulation’ published in 1975 in the Journal of Political Economy, argued that safer vehicles meant it was potentially less costly for an individual to be involved in an automobile accident. Firstly, mandatory safety regulations such as airbags and seatbelts, Peltzman postulated, would result in “drivers expending fewer resources to avoid being in an accident, and as such, the total number of crashes would increase.”

Secondly, Peltzman suggested that the behavioural change induced by mandatory safety regulations, could entirely offset any reduction in the probability of injury. However, even determining the probability of injuries is a complex matter. Because while more safety equipment might induce some drivers to behave less cautiously, the additional safeguards may prevent injuries that might otherwise have occurred.

Peltzman suggests that “there is some evidence that (auto-safety) regulation may have increased the share of any injury toll borne by pedestrians and increased the total number of accidents”. As a result of new regulations, the total number of injuries may decrease, increase…or stay the same. But, as prominent economist Gordon Tullock satirically posited, if a regulator’s prime target was to force people to drive more carefully and reduce the number of crashes, they would mandate a sharp spike on the centre of the driving wheel.

Fewer crashes, perhaps, but a much higher chance of death from any crashes that did occur. And yet, that improving the safety standards of vehicles will lead to more reckless driver behaviour and more crashes is just too hard to test in an ever-changing world. While Peltzman found early data to support his hypothesis, subsequent research has failed to reach a similar consensus. Or any consensus at all.

Mixed results of later studies show there is no definitive answer. More importantly, they illustrate the effects of complicating factors – like weather, road conditions, insurance, driver skills, vehicle quality, tyre conditions and sunstrike for example – that also come into play in vehicle accidents and injuries, unrelated to safety regulations that must be corrected for. As a result of these later studies and their conflicting findings, some policymakers might conclude that you can never prove the Peltzman Effect.

But this is where the aforementioned economists Sobel and Nesbitt designed an ingenious solution. In order to test the Peltzman hypothesis, the two economists looked at NASCAR (National Association for Stock Car Auto Racing), using data on injuries, cautions and accidents, speed, race distance, number of cars, and prize money for every racing season over a set number of years.

NASCAR provides an ideal case study for investigating the empirics around the Peltzman Effect since each vehicle fleet competing in a given race must comply with set standards and are considered identical as a result. Driving conditions are also kept as constant as possible given that NASCAR postpones races if weather is unfavourable. Furthermore, the quality of the race car drivers themselves is kept largely constant given that they are all highly skilled.

The use of NASCAR data was effective because it already corrected for myriad confounding factors that the Peltzman study had not accounted for. Sobel and Nesbit found that while NASCAR drivers do indeed drive more recklessly as measured by the number of accidents and cautions, the likelihood of driver injury has fallen. That is to say, that while improved vehicle safety has led to an increased number of crashes, the probability of death and injury has decreased.

This experiment broadly confirms the Peltzman Effect, however it must be treated with caution. While in this case, safety regulation around vehicles induce riskier driver behaviour, this does not mean all safety regulations or technological advancements should be shunned in favour of spikes on steering wheels. Improved safety regulation have saved more lives as illustrated by the NASCAR case. And NASCAR spectators like crashes more than most people do. When confronted with highly complicated problems where reality and the whiteboard appear to be in conflict, it may seem much simpler to relegate the issue to the ‘too-hard basket’.

But Sobel and Nesbit illustrate how smart economists can find novel ways around difficult public policy problems. The second lesson from Sobel and Nesbit is more salient. The NASCAR experiment illustrates the need to be cognisant of offsetting behaviours and unintended incentives that a policy might create.

Most importantly, policymakers must ensure that the benefits of statutory change exceed the unintended costs of potential offsetting effects.

---------------------

*Khyaati Acharya is a research assistant at The New Zealand Initiative.