Publications                   Bio                     New and Extra Publications                


Much on airbags in Chapter 11, Chapter 12, and Chapter 15 of 

Traffic Safety (Copyright © 2004 by Leonard Evans) 

This appeared, with minor variations, under the following titles

Offering motorists the airbag option.  Washington Times, 8 June 1997

Air bags -- Even the smartest technology is a dumb, dangerous mandate.  Detroit Free Press, 16 June 1997

Air bags have been oversold.  Plain Dealer, 16 Nov. 1997

Since then the evidence of the correctness of the views expressed has continued to mount –  see Traffic Safety (Copyright © 2004 by Leonard Evans) - but nothing has substantially changed

Mandating airbags -- wrong then, more clearly wrong now

Government airbag policy looks more and more like the emperor with no clothes.  Increasingly bizarre pronouncements are offered to conceal its nakedness, including:  Don’t place infants in front seats.  Don’t allow children under twelve to sit in front seats.  If you are short, move your seat back and add blocks to the pedals so you can still reach them.  Such actions are not merely inconvenient, they can reduce safety.  Infants and children in rear seats divert driver attention from the road ahead.  By moving seats back, short drivers reduce their already compromised forward view, and make braking more difficult.

The mandate requiring airbags was supported by government claims, documented in the 1977 Federal Register, that they would save over 90,000 lives per decade  --  an absurd claim relative to technical data then available.  The current (far too optimistic) claim is that they have saved 1700 lives from 1986 to 1996, less than two percent of the original claim, and an even smaller percent of the over 400,000 traffic deaths that occurred in this period.

The three best technical studies consistently find that airbags reduce the risk of death in a crash by 9% for belted drivers.  To interpret a 9% effectiveness, think of 100 belted drivers killed in cars without airbags.  If the cars had airbags (everything else being equal), then 91 would still die and 9 would survive, usually with severe injuries.  I am unaware of any intervention approved by the FDA with an effectiveness as low as 9%.  Even for those driving without safety belts, which is illegal in all states except New Hampshire, airbags reduce fatality risk by only 13%.

The claim that airbags have saved 1700 lives is an inference calculated from estimated effectiveness – the technical community cannot identify even one individual whose life was saved by an airbag.  In stark contrast, there are already 62 known individuals who were killed by airbags in crashes that, without the airbag, were unlikely to injure.  The vast majority of the claimed reduction of 1700 fatalities were unbelted males, nearly half of them drunk.  The government has chosen to not publish any estimate of the reduction (or is it an increase?) in the number of fatalities to sober belted women.

The claim that airbag policy has saved lives is based on a fundamental assumption that is false – namely, that the policy has zero influence on driver behavior.  A number of short ladies have informed me that they began to drive more cautiously when they realized airbags could kill them.  The opposite effect also occurs  --  the belief that we cannot get hurt produces less cautious driving.  If the misconception that airbags substantially reduce risk led to even a 2 mph increase in average travel speed, this would more than cancel the modest benefit of the airbag, thereby increasing national traffic fatalities.  Since 1980, traffic fatalities in Britain and Canada, countries where safety policy has focused on important items like drunk driving and belt wearing rather than on airbags, traffic fatalities declined by 39% and 40%, respectively.  If fatalities in the US had declined by 39% since 1980, we would now be experiencing over ten thousand fewer deaths per year.  Instead we are congratulating ourselves on saving 1700 lives in a decade, a claim that isn’t even true.

The most crucial problem with the airbag is the fundamental ethical one of killing people in some categories in order to save people in others.  There is now clear evidence that airbag effectiveness is lower than average for many large sectors of the population - females, short people, old people, and occupants of smaller cars.   Given that the effectiveness is 9% for the entire population, it cannot decline much before becoming negative.  It is therefore possible, and indeed likely, that the airbag increases the average risk of death in a crash for persons in broad categories containing millions of Americans.

While depowering can decrease the number of people killed by airbags, it will decrease even more the number saved, thus reducing the 9% effectiveness still further.

The problems with airbags are intrinsic – deploying an occupant protection device after the crash has already occurred is simply too late.  The problems cannot be solved by so-called smart airbags.  Improvements as large as 10% rarely occur in mature technologies; airbags have been around for over three decades.  Even if a somewhat implausible improvement in performance as large as 10% could be achieved, this would raise the present 9% effectiveness only to 9.9%.  Injecting a new generation of untested airbags under current regulations is tantamount to again compelling the American public to be involuntary guinea pigs in an experiment in which we already know millions will be placed at increased risk of being killed.   It is simply unacceptable to turn the old rule of the sea on its head, and make it “women and children first” to be sacrificed in order to protect large unbelted men.

Airbags were mandated because they were supposed to be passive devices requiring no user actions.  In fact they are far less passive than manual safety belts, which require users to learn only one simple rule  --  fasten your belt.  Doing so reduces driver fatality risk by 42% without negative side effects.  Given the danger, inconvenience, cost, burdensome instructions, and low effectiveness of airbags, their mandatory installation would never have been considered if the airbag were merely a technical device.  It has, however, acquired the status of a religious icon to the anti-technical believers who inflicted the mandate on the nation in the first place, and whose beliefs are too strong for them to admit that they were wrong.

The only reasonable approach is for the Government to neither prohibit nor mandate airbags or airbag on-off switches for the new or after-market.  If consumers are allowed to choose, some will make bad choices.  This is still far preferable to the present situation in which the government compels bad choices on millions of us.



Leonard Evans

973 Satterlee Road

Bloomfield Hills  MI 48304-3153