Problem: SUVs and light trucks are booming in popularity. Because these vehicles have limited visibility in the rear, parents are crushing their kids while backing up out of the driveway.
Solution: A government mandate for rear backup cameras. As usual, this will be a one-size-fits-all regulation, applying to small cars too — even though the problem is mainly limited to hulking SUVs and light trucks:
When pickups and multipurpose passenger vehicles strike a pedestrian in a backover crash, the incident is four times more likely to result in a fatality than if the striking vehicle were a passenger car.
But the real question will be whether this regulation will be another test case for the theory of risk compensation. Proponents of risk compensation argue that drivers accept a certain level of risk. Rather than use safety devices to minimize risk, they instead use them to drive more aggressively. For example, exploiting anti-lock brakes to go faster down a snowy highway. This behavior has been observed on almost every safety innovation, even safety belts and motorcycle helmets.
The theory of risk compensation predicts that backup cameras will at best have no affect, and at worse cause an increase in fatalities. Once the camera shows the “all clear”, will drivers still take the time to look back and properly assess their surroundings? Or will they just go flying out of the driveway?
Systemic Failure 
[…] the U.S. DOT has decided to mandate rear-looking cameras in all automobiles [PDF]. Network blog Systemic Failure wonders whether this new regulation will give motorists the information they need to drive more […]
Hmm, in your last post on risk-compensation theory (RCT), here, you point out that RCT discusses risks to drivers, not to other people around the vehicle.
The eight studies cited in Wikipedia’s risk compensation article—seat belts, ABS, skydiving, motorcycle helmets, bicycle helmets, ski helmets, speed limits, safety equipment for kids, and condoms—all discuss risk to the individual using the safety equipment, not to other parties involved.
Are backup cameras a novel application of RCT or what? Please explain.
Jonathan,
Dr. Adams argues that other people are put at greater risk. In particular, ABS, seatbelts, and other devices encourage faster and more reckless driving, thereby putting non-motorized road users at greater risk.
The Adams quote about seatbelts implies that the mix of accidents shifted from inside the vehicle to outside the vehicle, while the total number of accidents remained the same. It’s unclear how many people die inside vehicles while backing up (into giant car-crushers, maybe), but I don’t think it’s very many. I don’t see how this can be a test case for risk-compensation if the people performing the action (the ones inside the vehicle) aren’t taking any risks to themselves while using the feature.
For some reason I found your blog two years later, while reading about this topic in relation to the legislation being delayed this week. The very first thing that occurred to me was risk compensation.
I think that the fact that backing up can’t injure the driver is not important. The potential damage to ones’ life (family,lifestyle, future, income, and so on) from killing another human being by accident is enormous. I think most people consider risks such as going to jail, destroying their family, and so on as equally, if not more, important than the risk of simple physical harm to themselves.
I doubt the cameras will have much effect unless they include active avoidance systems (e.g. can detect objects in the path of the vehicle and will prevent the driver from proceeding). In fact I wonder if it won’t actually make things worse.
We’re talking about hundreds of millions of uses every single day – where people become used to looking at a tiny video pic INSTEAD of looking around physically. I suspect that many of these accidents occur not because of blind spots, but because of negligence. The camera falls squarely in the arena of things that give you a false sense of security: risk compensation. Craning your neck is harder than glancing at the camera. Doing both would be the best thing, do we really believe the average American will be interested in adding a step to a routine, rather than trading one for a simpler one?
This data should already be available. I am surprised nobody has done a study of backover rates in camera-equipped vs. non-camera-equipped vehicles. There must be enough data by now to do a good one.