Shifting Risk Instead of Reducing Risk

Risks of teen driving:

For more than a decade, California and other states have kept their newest teen drivers on a tight leash, restricting the hours when they can get behind the wheel and whom they can bring along as passengers. Public officials were confident that their get-tough policies were saving lives.

Now, though, a nationwide analysis of crash data suggests that the restrictions may have backfired: While the number of fatal crashes among 16- and 17-year-old drivers has fallen, deadly accidents among 18-to-19-year-olds have risen by an almost equal amount. In effect, experts say, the programs that dole out driving privileges in stages, however well-intentioned, have merely shifted the ranks of inexperienced drivers from younger to older teens.

Posted on September 21, 2011 at 6:58 AM57 Comments


Eric TF Bat September 21, 2011 7:19 AM

In Australia, the demonising of P-platers (drivers in their first couple of years of driving) is driven almost entirely by the Murdoch press, who seem to like hating on the youngsters because their target market are old, stupid and bitter. P-platers are initially restricted to drive at no more than 80 km/h even in 100 or 110 zones (I’d translate that to mph for you, but come on – it’s the 21st century), which basically means they provide moving barriers on every highway and freeway to increase the chance of accidents. And I’d always suspected that the result was a bunch of drivers who are supposedly trained but who have no idea how to handle cars at higher (legal) speeds, so this article is thoroughly unsurprising.

Mike B September 21, 2011 7:22 AM

Well you can’t say it wasn’t worth a try. I suspect the safety benefits were undermined by teens simply delaying their entry into the driving market instead of going through the whole graduated regime. A better law might simply institute either a graduated privilege or a more challenging test for ALL new drivers.

David September 21, 2011 7:38 AM

Continuing Eric’s comment… the latest lunacy here in Australia is to require all Learner Drivers to undergo 120 hours of log-book certified traineeship under the supervision of an experienced driver (probably their parents or an older sibling).

The problem on the roads is that whereby in the past when we saw an L-plater, we knew they were very inexperienced and should be given a lot of latitude (both physical and mental!) now we have no idea whether they at hour 1 or hour 101 of their progression. Worse, poorer families are complaining that they can afford neither the time nor the fuel to complete the 120 hours (if indeed they own a car!). Hint: cheating is rife! And entirely undetected.

So, I’m probably also disagreeing with Mike’s graduated regime too.

Once we agree on a set of rules that don’t require a ‘pass’ or ‘fail,’ but instead assess as ‘competent’ or ‘not yet competent’ we will permit all qualified drivers to become drivers, not proto-drivers.

ZG September 21, 2011 7:41 AM

Its a pointed example of how we shelter our children from all kinds of danger and expect that they will handle it ok when they are older. Life is risky. The best we can do in all cases is to provide the best guidance we can and let them have the freedom to learn how to deal with the world on their own. Experience is always the best teacher. Where I work we have had some parents trying to attend job interviews with their children. That seems absurd to me but it is a symptom of a bigger societal problem.

Martin Budden September 21, 2011 7:44 AM

The LA Times article says:

Overall, since the first program was enacted in 1996, graduated programs were linked to 1,348 fewer fatal crashes involving 16-year-old drivers and 1,086 more fatal crashes involving 18-year-old drivers.

So not all the risk is shifted – there is also some reduction of risk. Overall it seems like almost 300 lives have been saved.

5b854efce19e3663b5f9209e493445dabf0a1efd September 21, 2011 7:49 AM

@Eric TF Bat

If you’ve already got large vehicles and trailers at a reduced speed limit in the first lane putting new drivers there too doesn’t make much of an obstruction. I don’t expect new drivers to like it though.

Arkh September 21, 2011 8:22 AM

@Martin Budden How much of this decrease has to do with better and more secure cars, better road equipment, faster safety teams intervention?

Muffin September 21, 2011 8:23 AM

@J: 25? You honestly want reckless young greenhorns like that on the road? I say restrict driving to those above 40! (Did I mention I’m 43 myself?)

Arkh September 21, 2011 8:24 AM

@J and people older than 25 should be prevented to drive as their mental and physical abilities are already declining.

Jamie McCarthy September 21, 2011 8:33 AM

The article makes clear that 80% of the fatal crash reduction is “made up for” by increased fatal crashes in older drivers.

But that means the teen-driver laws are reducing crashes. Only 20% as many as it might appear at first blush, but still, reducing crashes.

Shouldn’t the title be “Shifting Risk and Reducing Risk?”

If someone proposed a law that promised that the 4,054 annual teen driving fatalities could be reduced, even by a paltry 262 lives, wouldn’t that be worth considering?

Nigel September 21, 2011 8:55 AM



Also, I see drivers of all ages doing things like steering with their knee while holding a beverage in one hand, fixing their hair with the other hand, and reading material propped on the steering wheel.

And don’t get me started on the 95 year old ladies, faces two inches behind the steering wheel, going 60 kph in a 100 kph zone.

Geoffrey Wiseman September 21, 2011 8:57 AM

Ontario (Canada) has graduated licensing regardless of age; it sounds like some of the restrictions in California apply equally to a wide array of teenage years, so I’m not sure simply delaying license acquisition explains these stats.

Thomas B. September 21, 2011 8:59 AM

This whole driving thing conjures a bit of spartan imagery. Yes, many teens will die in the process of learning how to drive… but those who live on will be stronger!

Thomas B. September 21, 2011 9:00 AM

Also, where is Google with those self-driving cars?

I know a lot of folks who are leery of the idea, but ask them when their kids start driving, and they warm to it pretty quickly.

Franky B. September 21, 2011 9:00 AM

@Mike B:

I certainly CAN say that it wasn’t worth a try, because I predicted this exact outcome. The whole premise of a graduated system for youngsters is that their high accident rate is caused by immaturity. I’ve always posited that it’s rather caused by inexperience. This seems to correlate to that.

By shifting the age where drivers get to try more dangerous scenarios, you’re simply making it take longer for drivers to acquire said experience. Plus, adults tend to learn more slowly!

Z.Lozinski September 21, 2011 9:12 AM

Motoring accidents and the associated risks are very politicised. There is an interesting report from the UK’s Institute for Advanced Motoring (published April 2011) which analyses the contributory factors from 700,000 road accidents in the UK from 2005-2009.

Many of the issues which receive the most media coverage are not actually among the most common contributory factors. Speeding, drink driving, mobile phone use, tailgating, road rage and bad weather are all important but are not as frequently reported as driver errors;

BruceH September 21, 2011 10:21 AM

I think the lesson here is that driving is a skill. People who are learning to drive get into more accidents. Perhaps maturity does play a role, but I suspect that even if you increase the legal driving age to thirty, you’d find that 30-32 year-olds get into more accidents than the older population. Perhaps not at as high a rate as teenagers, but still a statistically higher rate than the general population.

Which is to say that while the proper age at which to issue permits may be debatable, shifting it by a couple years doesn’t do much to rectify the problem of higher accident rates for new drivers.

vlion September 21, 2011 10:26 AM

Meh, unrestricted access at 16.

Either we invent goldbergesque rules that get broken where the system has crazy bugs in it, or we don’t. Either way, accidents are still going to happen, morons still going to be morons.

Better to have simple systems that are not buggy and are easy to understand.

karrde September 21, 2011 10:28 AM

@BruceH:I think the lesson here is that driving is a skill. People who are learning to drive get into more accidents. Perhaps maturity does play a role, but I suspect that even if you increase the legal driving age to thirty, you’d find that 30-32 year-olds get into more accidents than the older population. Perhaps not at as high a rate as teenagers, but still a statistically higher rate than the general population.

Which is to say that while the proper age at which to issue permits may be debatable, shifting it by a couple years doesn’t do much to rectify the problem of higher accident rates for new drivers.

It’s not the years, it’s the mileage.

Captain Obvious September 21, 2011 10:31 AM

@Mike B

I suspect you’re right. Most drivers begin driving to go out with their friends or to taxi their younger siblings.

As these activities are banned until older, there’s little reason to start until then.

X September 21, 2011 10:35 AM

At 16 kids are still at home, attending high school, living with their parents.

At 18, they’re off at college, thousands of miles away from their parents.

Which provides better oversight?

(You’ll note the same applies to consuming alcohol…)

Doug Coulter September 21, 2011 10:57 AM

Anecdotally, I know only a few people who did not have a medium bad accident in their teens. I know I did. While it’s hard to remember in the mists of time, I recall that once I got “good” at it, I got kind of cocky — and a few near-misses only made me more confident that with my quick reflexes I could continue to avoid trouble. Nope, one time looking where my girlfriend pointed instead of at the road and bam. Luckily no one was seriously hurt (but both cars totaled).

Changed my life, no accidents my fault ever since, and millions of miles later (I did field service for a long time and drove a heck of a lot).

I swear, you should try kids in perhaps some simulator that could pull enough G’s to let them realize just how much energy is involved in a crash…or something that gets that across at the bloody nose or busted teeth level (there weren’t shoulder belts when I learned)…but whenever I think “there ought to be a law” or anything like that, I am humbled by the number of times those things act directly opposite to what’s hoped. So, forget it.

Anon September 21, 2011 12:01 PM

I don’t have much time to dive deeper but I’m guessing a laundry list of confounds could be brought to the table. I suppose I could read the entire paper before saying that but studies like these usually are very inconclusive.

Irene September 21, 2011 12:03 PM

Something we can never learn is whether the distribution of accidents is random, or is it the same set of teens (who tend to take more risks, or some other non-random factor) are now killed at 18 that would otherwise be killed at 16.

mcb September 21, 2011 12:55 PM

I’m not as worried about fatalities to these young drivers as I am about the fatalities, injuries, property damage, and traffic disruption caused by them. They can inflict all sorts of mayhem – and survive – if mommy has a big enough SUV equipped with the most modern supplemental restraints. To that end I wonder if the statistics are skewed according to the income of the persons purchasing the vehicles these beginners are crashing.

George September 21, 2011 1:34 PM

Yet another example of what happens when the targets of well-intended legislation fail to behave the way the well-intended authors anticipated. In this case, it’s important to find some way to punish the rebellious teens for their failure to make the graduated licensing scheme successful.

I’m sure those well-intentioned authors are at work right now coming up with additional restrictions and bureaucratic hurdles that will fix the problem. And if a few years later, the researchers find that it hasn’t fixed the problem, they’ll just add more restrictions and hurdles and repeat until the desired result is attained.

(That’s the TSA’s security model. I think everyone agrees that after a decade of continually adding restrictions and hurdles it has succeeded at creating an impenetrable bulwark against the terrorist threat.)

MarkH September 21, 2011 2:06 PM

I read the article when it first appeared … as a few commenters have noted, age-based driving restrictions are associated with a non-trivial reduction in fatal accidents.

Given this (and IIHS not finding such a shift), the rather negative tone of the comments in the article seemed rather puzzling to me — especially the final quote saying “the programs didn’t work,” when the data suggests that they may well have saved lives.

As to the interesting question of how much total reduction in fatal accidents is due to other factors, it should be possible to control for this to some degree, by comparing data from other age groups, or places with comparable road and vehicle makeup that lack such programs.

But some of the folks quoted in the article, and several of the commenters here, seem to simply accept that age-based driving restrictions haven’t worked.

It seems to me that the great majority of motorists are ignorant of many basic facts concerning road safety. But because so many people drive, hundreds of millions imagine themselves to be more or less expert on the topic. It is another example of a topic Bruce frequently illuminates: how poorly our minds our adapted to the assessment of risk in the modern world.

Daniel September 21, 2011 2:07 PM

I find these articles amusing because I started driving when I was 11. By age 13 I drove my younger brother and myself back and forth to school, a 60 mile round trip. The only ticket I ever got was a parking ticket because my mother failed to inform me about the parking laws when I first started driving.

Personally, I have always felt the right answer to the issue has been to get kids driving /sooner/ rather than later. When a kid is 10-13 they are still impressionable and can learn. By the time they are 16 they know everything already oh, yeah, and FRIENDS.

Jeremy September 21, 2011 2:29 PM

Article: “Overall, since the first program was enacted in 1996, graduated programs were linked to 1,348 fewer fatal crashes involving 16-year-old drivers and 1,086 more fatal crashes involving 18-year-old drivers.”

@Martin Budden: “So not all the risk is shifted – there is also some reduction of risk. Overall it seems like almost 300 lives have been saved.”

From the phrasing of the passage you quoted, I would guess those are cumulative totals, not annual. Which means you need to take into account that the accidents were also DELAYED 2 years by requiring the drivers to be 2 years older. Put another way, the data set includes 15 years (1996-2011) where ages 16-17 were operating under the new restrictions, but only 13 years where ages 18-19 were.

1348 deaths over 15 years is about 90 per year

1086 deaths over 13 years is about 84 per year, which is lower, but suggests more like 80 lives saved, not 300

Of course, it’s always possible I’m misunderstanding the data.

ZG September 21, 2011 2:40 PM

One of my all time favourite stories is from the “How to Keep Your Volkswagen Alive: A Manual Of Step-By-Step Procedures for the Compleat Idiot” written by John Muir, engineer, mechanic and philosopher. In the repair guide he tells a good story about his son. When he turned 16 he was given a VW Van, $50, some tools and a draft of this guide. His Dad told him to drive around the country and live life then come back. At first the son called often for help but in time he called less, he made it around the country and learned a lot about life. This may be fiction (or not) but it sounds like a good idea to me.

Jon September 21, 2011 3:15 PM

@ZG Hear Hear!

I got my first VW (a bus) at 16, and a copy of that beautiful book. I learned to drive with her, including sliding around in the snow. This paid off when, ten years later, I got to do some serious emergency sliding around obstacles suddenly appearing on the freeway.

But my favourite quote from “How to … Alive” was also about the bus, and it was:

“If we all drove as if we were strapped to the front of our cars like Aztec sacrifices, there would be a lot fewer accidents.”


Some random guy September 21, 2011 5:30 PM

I’ve never gotten myself into an accident and I’ve been driving since I was 16 (now 25, so I’m still a young’n). However, I have been a passenger in at least 5 car accidents throughout my life, and I know how easy it is for them to happen from just a moment of carelessness.

I think someone should make a simulator which attempts to check whether or not someone is paying attention and tries to distract the driver. When the program thinks the driver isn’t paying attention anymore, it makes a situation occur as quickly as possible and BAM accident out of nowhere (whether it be a deer, a sea of brake lights, pedestrian jay walking, black ice, oil slicks, etc.). The simulator could have some kind of hardware (a structure that basically contains at least the driver’s seat of a car) which simulates an actual collision by shaking or something (with a realistic amount of force). Even if the driver doesn’t lose focus, random events that could cause accidents should happen at random. If the driver manages to avoid getting into an accident, it will probably drive home the point that paying attention is crucial to driving safely, even when no other cars are around.

jb September 21, 2011 5:41 PM

We have far more learner and probationary drivers than reduced speed vehicles over here.
Most trucks will be on or just over the speed limit, reduced speed vehicles will be very few and far between and only under by maybe 10 kmh, not 20-30.
Also, it’s not only on multi-lane carriageways.


Mr. E September 21, 2011 7:19 PM

This is probably the first time I’ve read a thread on Schneier’s blog and thought, “What a collection of thoughtless comments.”

For those who insist that the risk has exclusively been shifted, and not reduced (in spite of the evidence that has been suggested by more than one poster already), consider the consequences of manipulating the driving age in the other direction.

Would 8 year old drivers be involved in as many injury/fatality wrecks as 18 year olds? How about 6 year olds? We can outfit them with special equipment if they can’t reach the pedals.

Since we’re only shifting risk, the numbers should remain about constant. Immaturity? Pish posh! Hours behind the wheel is all that counts.

mcb September 21, 2011 9:47 PM

@ John E. Bredehoft

“Legislators, who are primarily influenced by the attitudes of their constituents, are not necessarily persuaded by statistical evidence.”

Unless the statistical evidence is in alignment with the attitudes of the legislators’ constituents.

Miksa September 22, 2011 4:22 AM

Would more extensive driving training be better solution and would this current system provide an opportunity for creating such a training program. It seems mandatory extensive training would be politically impossible, but maybe with the current system young drivers could remove the restrictions if they go to a driving school. That could also earn them reductions in insurances. Here’s an example of the training programs use in Finland.

MRC September 22, 2011 5:19 AM

Want to know the biggest entry bar into driving? Cash.
Simply make cars too expensive for the average under 30 year old to afford and add rules making it illegal to sell any car without a huge amount of safety features.
Then sit back and watch the driver deaths plummet.
Of course this is achieved by severely reducing the numbers of people who can drive in the first place, and probably an increase in cyclist deaths as well

Rhialto September 22, 2011 7:05 AM

Mr E: I would go even further! Let the kids start driving at age -2, so they get all these accidents before they’re even born. That will take care of the whole issue.

spamse September 22, 2011 7:49 AM

The death statistics don’t discriminate between the driver/passengers and third parties. It’s possible that the 16-17 statistics represent a large number of third parties being killed due to inexperience/incompetence of the driver. Perhaps the 18-19 accidents are more weighted towards occupants of the car – not through incompetence, but disregard for the law or misadventure. This is, of course, mere speculation,but I think it’s fair to criticise conclusions made from such broad statistics.

It’s certainly hard to save people from themselves (particularly young males), but saving other vulnerable road users from them is no bad thing.

LinkTheValiant September 22, 2011 8:21 AM

Why NOT at least begin training younger children to drive, even if they don’t actually use the skill until later in life? The concept of a child turning 16 and receiving the ability to drive is ludicrous. You wouldn’t do that with a firearm, or a table-saw, or an industrial shredder.

In fact, driver’s education ought to be part of standard curriculum in middle school. Children should be learning a skill which is useless to them at the time, but will stand them in good stead later. (After all, most high school students don’t have a direct use for chemistry or biology, yet the subjects are still taught.)

I was operating farm equipment when I was eight, so I understood the principles of driving long before I was legally able to operate a motor vehicle. I had several accidents involving tractors during this time, all of them due to carelessness. But because tractors typically move at approximately 5-10 MPH (8-16 KPH for the rest of the world 😉 the only real effect of the accidents was property damage.

No, I didn’t enjoy fixing what I broke, but it certainly gave me a respect for driving that was MUCH less expensive than a human life.

squarooticus September 22, 2011 11:16 AM

MRC: Systems like ZipCar provide hourly rental wheels to those who can’t afford to own them. You’d have to make cars awfully expensive to make this business model impossible.

John September 22, 2011 12:32 PM

I remember seeing a similar finding related to moving the drinking age to 21. That is, the same number of drunken-driving incidents, just from older people.

Dirk Praet September 22, 2011 6:55 PM

Strange that nobody has brought this up yet, but mitigation in risk management is not just about blind staring at one aspect of the issue, but just as well about considering alternative solutions and thinking “out of the box”.

If reducing car injuries and fatalities is the purpose, this can also be achieved – and for all ages – by providing and promoting ubiquitous, affordable and on-time public transport systems. A nice plus would be the benefits to the environment, a decreased dependance on oil and a firm middle finger to Big Oil’s influence on politics and society as a whole.

GreenSquirrel September 23, 2011 3:55 AM

@Jamie “If someone proposed a law that promised that the 4,054 annual teen driving fatalities could be reduced, even by a paltry 262 lives, wouldn’t that be worth considering?”

not really no.

But I do know a law that could reduce ALL driving fatalities to zero.

Would anyone want to consider it?

Christopher September 23, 2011 6:29 AM

If you drive for work, it should be tied to a work permit, ergo if you are 15 and actually got a job delivering pizza, great, it’s on you and the pizza joint owner to be responsible behind the wheel.

At eighteen, all restrictions come off. For everything, including drinking. If we can send you off to die for Big Corp, then we shouldn’t keep you from a cold Bud on your first rotation home.


Clive Robinson September 23, 2011 7:36 AM

I hate to disagree with many people here but…

In most cases it’s not the nominal age difference, and it’s not the mileage covered, it’s something we have known for years but chosen to ignore for “political reasons” and “down right stupidity”.

Driving is a “mechanical skill” requiring the use and good coordination of all limbs, eyes and ears. In this respect it is like playing a complex sport or musical instrument.

When we learn a sport or instrument we are not thrown in at the deep end with proffesional players, we are taught by graded exercises the skills we need. We are then expected to devote a considerable time in repetative practice to master each additional skill, usually using contrived practice to enhance the learning of the particular skill.

So how do we currently teach people to drive?

Simple answer “in about the most stupid way possible”, that is we chuck them in at the deep end on the public highways and hope they learn enough in a few short hours about how to control a complex and extreamly dangerous tool to not hurt themselves or others…

In this respect it’s a bit like learning to do a “unicycle knife throwing act in a crowded auditorium”. I’ts far from ideal, especialy if you are a member of the audiance who gets an accidental miss and loses a bit of their anatomy for the benifit of the budding ride by knife artists training.

The reason for this stupidity originaly was that the roads were a lot safer in the past and there realy was no other way to learn to drive. The reason the stupidity continues today is mainly political, and involves “vested interests”. In many places day to day living at what is considered a normal level is all but impossible if you cannot drive, thus driving is treated by the populous as an essential right in those and many other places.

Further there is a whole industry built up around this stupidity and no real desire by either the industry or the politicians to change it.

Not so other forms of transport such as proffessionaly piloting an aircraft or large marine vessel. Because of the costs involved much of this training is done in simulators these days and we do have trained “zero air/sea hours” pilots as a reality.

Whilst I’m not suggesting that we should switch over night to simulator only training, I would sugest the increased use of “hazard perception” training in simulators would give learners those skills more safely. Some studies sugest that “computer games” with even semi-realistic controls enhance hazard perception (no I don’t want to think what a “zero road hour” driver trained on “Grand theft auto” would be like).

Also I notice that nobody has mentioned the inverse corelation between safety features and things like speed (although the refrence to “Aztec Sacrifices” is a good analogy).

The small drop in the number of fatalities (after normalisation for time spans) in the younger group could be due to the fact that second hand cars available today have more safety features than 2 years ago and likewise for every year preceding it for the past decade or so. And that many young drivers don’t buy their choice of first car that is their parents buy it for them and have significant curtailing on “desire” (as does the insurance cost). Also these days (in the UK) the “kids” keep their first car until they can aford to upgrade to the car they want / desire.

However there is a down side to safer cars, as humans we all have risk threasholds that wildly differ from person to person which is why we have “extream sports”. But what we do know is our perception of risk is generaly limited to “me and mine” that is ourselves and those known well to us (family and friends). Further we also know that at a visceral level we have little or no regard for those not in the “me and mine” group and actually usually regard them as impediments or threats to our own desires or wishes.

Making a vehicle safer for the occupants makes this situation a lot worse, because it alows the occupants (me and mine) to significantly increase their speed etc by the amount of increased safety the vehicle gives, and thus makes it considerably more dangerous for those not in the vehicle (ie others unknown to the driver).

But when you look at it a bit further you find most drivers perception makes their increase in speed linear with the percived increased level of safety not the reality of the level of harm which is based on the impact energy of velocity squared.

Why? Even though it should be demonstrated each time a driver puts their foot on the brake in terms of stopping distance, the reality is that due to the vast improvment in tires and brakes stopping distances at speed are comming down a lot. Which is a problem because the energy that gets delivered to the object a half ton vehicle moving at 30mph hits on impact is the same irrespective of the efficiency of the braking system the vehicle has, and at the end of the day it’s this energy that kills maims and disfigues.

With regards to age, yes the younger you are the faster and generaly better you pick up a skill. We know from racing drivers that starting to drive go-carts competitavly around the age of eight is advantageous.

We also know from some schemes organised to stop joy riders that teaching those from 10-15 how to build maintain and race go-carts, stock cars etc actually reduces significantly their redoffending rate (and the danger they present to others and themselves).

As for mileage, well you can drive a thousand miles on long straight empty roads or round an empty track and learn little or nothing, simply because you don’t get into a chalenging situation from which you learn. Likewise taking a short drive in some places (Paris and Milan spring to mind) can teach you a great deal very quickly because of the shear number of challenges it presents.

r cox September 23, 2011 9:36 AM

While there does not seem to be decrease in fatalities for a specific age group, there has been a decrease in traffic fatalities overall. While it would be difficult to relate this decrease to the teen driving laws, it is no more difficult than attributing the deaths in young adult drivers as being shifted from the teen driver category.

Another thing to consider for those who are nostalgic is the shift in the number and power of cars. When I was a kid, it was unusual for a family to have more than two cars, many had one. It was not unusual for children and teens in the city to walk, bike, or use the bus. Now it is not unusual to see homes with a car for each driving individual.

Dirk Praet September 23, 2011 10:59 AM

@ Clive

“Likewise taking a short drive in some places (Paris and Milan spring to mind) can teach you a great deal very quickly because of the shear number of challenges it presents.”

Paris and Milan really are walk in the park compared to utter nightmares such as New Delhi or other big cities in Asia.

Randy September 26, 2011 2:40 PM

I teach my children in a graduated manner. First I take them in an empty parking lot and have them pretend it’s a road, driving on the right and missing the imaginary cars in the empty parking spots. Here they can safely get a feel for the acceleration, braking and steering. I even tell them to slam on the brakes once in a while…just to feel it. Once I feel they are ready, they graduate to less traveled roads with wide berms.

These laws are in place to help with two dangers with young drivers.

The first is simple experience. Here in Ohio, USA they can get their learner’s permit at 15 1/2 years old, but then have to drive with a parent or other adult in the passenger seat. It’s up to that adult to both teach and not allow the learner to get into a situation that is too far beyond the teenager’s experience.

And once the teenager gets the license at 16, the second part of the law makes it illegal to have more than one other passenger other than family members. That means the teenager can drive only siblings or one unrelated passenger. This helps to reduce the level of peer-pressure to do something stupid. And I was always hoping that the new driver’s younger siblings would always tattle if the new driver would do something crazy or dangerous.

I like these laws because they codify exactly what I wanted to enforce, so I can say “It’s the law, so I can’t let you do it.”

I also don’t understand how the death rate would have increased in the 18-19 year olds, unless they still don’t have enough experience or they are delaying getting their license.


bob September 27, 2011 10:05 AM

During the Vietnam war, we started with an abysmal rate of losses in air combat. It turned out that the first 5 air-to-air combats you are in were likely to be fatal. Once you get to #6 you have enough experience you are more than likely going to live out the war.

The DoD decided the solution was to make a simulation so real that your first five AA combats essentially occurred before you ever left the country. They called it “Top Gun” (actually fighter weapons school or Red Flag, but more people are familiar with the movie name). Apparently they were correct because the kill ratio went from 1:1 to more like 10:1.

I suggest if you REALLY want to reduce fatalities, make a requirement that new drivers are REQUIRED to take (and pass!) an extremely rigorous driving school, and that an integral part of it be driving many hours in painfully realistic simulators where when you crash you can hit “reset” and try it again instead of dying. This way you can learn timing and vehicle response as well as ways to handle “bad” scenarios with STEEP learning curves BEFORE you put the public in jeopardy.

Germany has a MUCH more rigorous driver’s education requirement than the US – its about the same level of training required to get a DRIVERS license in Germany as it is to get a PILOTS license in the US – thats how they have no overall speed limit on the highways in a country much more densely populated, yet have a LOWER highway fatality rate than we do.

The hard part would be that a lot of people could not pass something this rigorous and would be prohibited from driving. I like it – but its a tough political sell in a country where the economy is built around the auto industry and wants maximum crashes to keep people employed making, selling and repairing cars and all the dependent industries thereto.

Theresa September 27, 2011 11:25 AM

Thanks Bob for your wonderful piece on driver and pilot training. I have logged 2.5 million safe driving miles as a truck driver without any accidents. I attribute this to the stringent training I received in 1979 from the Teamsters Union. We were in training class a full month before being able to drive on public roads.

I have recently trained in a simulator to train for driving a Leibherr, a 400 tons haul truck used in gold mines. The simulation was a great experience in learning responses in real time.

Leave a comment


Allowed HTML <a href="URL"> • <em> <cite> <i> • <strong> <b> • <sub> <sup> • <ul> <ol> <li> • <blockquote> <pre> Markdown Extra syntax via

Sidebar photo of Bruce Schneier by Joe MacInnis.