*Foolproof: Why Safety Can Be Dangerous and How Danger Makes Us Safe*

The new Greg Ip book is coming out soon, I just bought my copy…


I know that authors don't always have full control over their book titles or marketing descriptions, and I also know that Greg Ip is fairly smart, sensible, and sober as an analyst, but it seems to me - just going off the abstract - that the book may be confusing between local and global optima. For example, helmets in American football are an example of local optimization, while banning/preventing head contact in sports would be an example of global optimization. The book seems at risk of conflating the two concepts - risk solutions which only reach local optima would indeed do nothing to alleviate exposure to more structural/systemic risk. There's also the issue of tradeoffs, in addition to unintended consequences

Improved helmets resulted in more not fewer injuries as players began using the improved helmets as battering rams to inflict injury on other players. My sister's husband played college football in the 1950s. I tease him that he played so long ago that he wore a leather helmet. Not quite, but he did play without a face mask, not because they hadn't been invented but because it took awhile before football adopted the rule (the "safety" rule) that players couldn't grab an opposing player by the face mask. As a running back, my sister's husband was far more concerned about a broken neck (caused by a lineman or linebacker grabbing his face mask as he ran through the line) than a broken nose or a few broken teeth or a damaged eye. At age 82, my sister's husband now suffers all kinds of physical ailments as the result of playing football all those years ago, including blindness in one eye (no face mask), no knee caps, and a spinal cord that must look like a pretzel. But the damage done to him years ago by football pales in comparison to the damage done to him in the past ten years by his doctors.

" .... pales in comparison to the damage done to him in the past ten years by his doctors." I have a friend whose life was ruined, in her twenties, by incompetent doctoring performed in Texas. As she says, she has plenty of criticisms to make of the NHS but none remotely as important.

Another issue to global/local optima, along the lines of what Andrew suggests above: suppose seat belts make drivers engage in more risky driving, and actually increase the accident rate. Then it would be foolish NOT to drive with seat belts, since due to the increased accident rate a person NOT wearing seatbelts is at greater risk. Hence laws requiring seatbelts be worn are sound.

I also propose--and will patent it soon--an auto safety mechanism consisting of a steering wheel with a dagger pointed straight at the heart of the driver, which will impale the driver in the event of the slightest accident. The theory is the driver will be more careful when operating the automobile.

The "dagger on steering wheel" idaea is quite old, and quite tired. The actual fact of the matter is that the number of road fatalities per driven kilometer has dropped for a long time and continues to drop. Like all forms of violence and accidents, the incidence rate is higher in the US than in the rest of teh civilized world, but it is still dropping. The idea that less safe cars would decrease the accident rate is not supported by data.

I think it's more an argument for scepticism about the benefits of some safety measures. After seatbelt laws are introduced, casualties consistently fall - but on the same trajectory they followed before the seatbelt law (Smeed's Law). Of course the overall story is lots of incremental changes to improve safety, so I wouldn't write them off as pointless. As an old-fashioned liberal, I'm happy with strict traffic laws firmly enforced when they put others at risk, but I'm less comfortable with laws making us do things for only our own benefit. There's much stronger evidence of the benefit of motorbike helmet laws. It really is suicidal not to wear one. But who am I to judge the risk of death against the benefit of wind through hair?

"The idea that less safe cars would decrease the accident rate"

That's not the point behind the "dagger on steering wheel" argument at all. The argument is that with safety equipment in general you have two opposing effects and it often isn't obvious before the fact which one, if either, will dominate. With any safety device like seat belts or football helmets, you get a benefit in any given impact, but that added protection may induce less caution and hence increase either the number or severity of impacts. Whether the added protection outweighs the lowered caution in terms of overall safety is an empirical question (that in the case of seat belts I believe is clearly positive).

Also, the incidence of impacts doesn't have to go up either (e.g. seat belts don't have to increase the number of accidents), but, as in the football example, the force of impacts may be greater, which might, for example, correspond in the auto case to people driving faster with seat belts than they would without. It may be entirely anecdotal, but if you took away the seat belts and airbags from my car today, there is little doubt I'd be much more cautious driving home tonight.

Your proposal nicely points out that seat belts do not "make drivers engage in more risky driving." Drivers assess incentives and choose more risky driving.

Here's a thought experiment. I am sure most people here will predict that what they *want* to happen *will* happen.

Abolish all speed limits. Keep radar equipment and record the speeds but not the registration details. Publish the speed at impact of every vehicle involved in a death or hospitalisation vs. average speed of all vehicles on that road.

That will have little influence when I am running late for work TODAY.

Alan, can you tell me the increased morbidity that results from smoking two packs a day without looking it up? Maybe you can, but most people can't despite that information being available and very relevant to the decision to take up or quit smoking. So I don't suppose people will pay much attention to published speeds of vehicles involved in fatalities or injuries. Except when they go to court.

If you are looking for other means to make money, come do trading like I did and you’ll see how profitable it is. Google Superior Trading System they’ll teach you how.

If economists ran the world, we would all still be walking. Riding animals requires too much dangerous research and development with no assurance of reward.

Sure, walking is slow and thus dangerous because you may not be able to outrun the bear or your enemy who did not listen to the economists and not only learned to ride horses but invented spears.

But if you did learn to ride horses and invent the spear or worse the bow and arrow, you will only do something stupid like invade your enemy's territory only to find he invented the shield and a longer spear which he uses to ward off your arrows and then spear you

Safer to never try to do more than walk. After all, running is more dangerous that walking, and the end we are all dead.

It's never been safer traveling from point A to point B then it is now in the US where literally millions of miles are covered per traffic fatality. But it's not because of seat belts, which aren't a safety mechanism but instead one of control. If safety were a real concern drive-through windows at fast food restaurants would be forbidden and driving around with a dog on one's lap would be illegal.

Interesting post, but I don't think you have a very good understanding of economics.

In the author's name "Greg Ip" what is Ip an abbreviation for ?

Please review it once you have it in hand. Thank you.

This seems to be much in line with Nassim Taleb's idea of anti-fragility and having "skin-in-the-game" based on what I can surmise from the reviews on amazon.

Best thing I've read about safety in the last day or so was a claim that the reason mandatory bike helmet use resulted in fewer injuries was due to it discouraging kids from riding bikes. That's like a piece of candy. You'd think it'd be reasonably easy to check (Are there fewer accidents?) but I hope no one bothers to.

The study data does not say this at all.

The one study on this claims a 19% reduction in certain kinds of injuries and a 4% reduction in bike usage. If the study is even accurate, this means only a small part of the injury reduction is due to reduced cycling unless you believe the people who stopped cycling were by far the most injury prone.

While there certainly is plenty of evidence for risk-compensation, many people with an ax to grind immediately make unfounded conclusions based on only partial information.

For example, in the debate on football helmets, stating the football helmets increase injury rates does not lead to any conclusion; the question is whether the degree it does this outweighs any reduction in catastrophic injuries. Also this is confounded with the rules of the game.

I had an idea for a safety pan -- forget it on a hot stove and it won't burn the food or catch fire because at a certain temperature it turns from a heat conductor to an insulator. I had a couple of ideas on how to make it work, but then I realized how dangerous this could be. If somebody grows up in a household with nothing but safety pans, then moves in with someone who uses regular pans, there's a recipe for a kitchen fire.

You learn the Martial Arts discipline from arunod the training,the better the discipline you have to the Martial Art the greater you become.Remember that this may be a very subconscious area for a lot ofpeople and they usually start working harder and concentratingmore without realising.The respect comes from the sparring, the harder youhit someone the harder they are going to hit you back, the more you respect that fighter themore they will respect you.

Comments for this post are closed