[pdonis]

The article quotes the 30,000 per year number, but it only really talks about fatalities or injuries to pedestrians by cars. The 30,000 number includes fatalities or injuries to people in cars as well. So a lot of relevant data is being left out.

Wikipedia has a link to NHTSA figures that give more detail. The link is here:

http://en.wikipedia.org/wiki/List_of_motor_vehicle_deaths_in...

From the 2010 data:

* 32,885 total fatalities

* 23,303 (70.8 percent) of those were vehicle occupants

* 4,502 (13.7 percent) were on motorcycles (drivers or passengers)

* 4,280 (13.0 percent) were pedestrians

* 618 (1.9 percent) were cyclists

* 182 (0.6 percent) were "other/unknown non-occupants"

So the article is focusing in on something that's, at most, only 15 percent or so (counting pedestrians and cyclists, since the article talks about both) of the problem. I say "at most" because of other interesting statistics from the NHTSA report:

* 31 percent of fatalities were in incidents involving alcohol-impaired drivers.

* 47 percent of the crashes that resulted in pedestrian fatalities had alcohol use reported by either the driver or the pedestrian or both. Interestingly, 33 percent of the pedestrians involved in these crashes were alcohol impaired, but only 14 percent of the drivers involved were; in 6 percent of the crashes, both the driver and the pedestrian were alcohol impaired.

* 32 percent of fatalities were in incidents where the driver was speeding. (42 percent of the drivers who were speeding were alcohol-impaired.)

* 42 percent of motorcycle drivers, and 51 percent of motorcycle passengers, who were fatally injured were not wearing helmets.

* 51 percent of vehicle occupants killed were not restrained (not wearing seat belts, or not in child seats/restraints).

* 11 percent of fatalities involved large trucks (gross vehicle weight over 10,000 pounds); of those, 76 percent were occupants of other vehicles, 14 percent were occupants of the trucks, and 10 percent were pedestrians.

Taking all this into consideration, I see a very different picture from "vehicles are murder machines". It looks to me like the major factors causing fatalities are individual choices made by people that put them at higher risk. And it's not like those choices are difficult, or about things that most people don't know. How hard is it to wear a seat belt? To put a helmet on if you're on a motorcycle? To not drink and drive?

[bruce511]

There seems to be some serious low-hanging fruit here with regard to improving these numbers. Imagine if you took the billions currently spent by the TSA and applied it to road safety?

Given what the rest of world does with airport security, it's possible to maintain airport security while at the same time freeing up massive resources which could be channeled into education (seriously, no seatbelts in 2014? ), policing of drunk driving, zero tolerance for speeding and so on.

Of course this won't happen. Regardless of the fact it is waay more dangerous to drive to the airport than it is to fly on a plane - it's easy to do a big song and dance at the airport to "make flying safer" (hint: it's exactly as safe as it is everywhere else where they don't succumb to the TSA rhetoric).

Americans don't care about the TSA at airports precisely because most Americans don't fly. So politically it costs no votes, (and with enough rhetoric some Americans even believe it's making things safer) whereas if you started creating or enforcing speed restrictions or drunk driving there'd be endless Libertarian arguments about the perils of big government, and the "freedom" of motor cyclists to not wear a helmet if they so choose.

There actually could be a really useful agency with the words Trasportation and Safety in the name, but it isn't the one that currently exists.

[pdonis]

most Americans don't fly

I'm not sure that's actually true. According to the Bureau of Transportation Statistics, there were 646 million enplaned passengers in 2013. ("Enplaned passengers" means passengers boarding originating flights, i.e., either nonstops to the destination or the first leg of an itinerary with at least one connection.) That's more than twice the population of the US, and about 2.7 times the population of the US that's over 18. Even allowing for the fact that many business travelers take many trips per year, there are still a lot of people who only take one or two airline trips per year, for vacations or special events, so it's hard for me to see how the percentage of Americans who don't fly can amount to "most"; I'd be surprised if it was half.

BTS link: http://www.transtats.bts.gov/

if you started creating or enforcing speed restrictions or drunk driving there'd be endless Libertarian arguments about the perils of big government, and the "freedom" of motor cyclists to not wear a helmet if they so choose

I think you're seriously underestimating the costs of enforcing driving regulations to the extent that you would need to to improve the numbers significantly by that means; I suspect that a more realistic estimate of those costs is what would lead most people (including me) to oppose such regulations. Whatever one's opinion about airport security, the amount spent on it is still a small fraction of government budgets and an even smaller fraction of GDP; and even for business travelers that take many trips per year, the overall cost in time of security screening is not that much compared to the time spent on the flight itself, particularly since most business travelers are easier to screen since they're familiar with the process and so get through it more quickly than people who only fly occasionally. (Note that I'm not arguing here that what is spent is justified by the benefits; I'm merely trying to compare the cost of airport security, in both time and money, with the cost of the sort of traffic regulation you're proposing.)

On the other hand, the kind of regulation you would need to have in order to, for example, make a serious dent in the rate of speeding, would be extremely draconian and would have huge costs. You would need traffic police on every major road, stopping people left and right; and you would need to do this continuously, or at least often enough that people would have a fairly high probability of getting caught if they speed. After all, most people do not speed just for the fun of it; they speed because it benefits them, because they get to wherever they are going faster. (I would also argue that most speed limits on highways are set too low, particularly given how the performance and handling of cars has improved over the years.) To make people not speed, you would have to raise its costs, in money and time lost, enough to outweigh the benefits, and that's a very tall order. Plus, you would have to hire all those traffic police, and pay them, and all the time they spend catching speeders would not be spent dealing with other issues that might actually be more important.

Similar remarks apply, although possibly with less force, to other types of regulations: for example, trying to enforce seat belt laws, or motorcycle helmet laws, or drunk driving laws. I say "possibly with less force" because the "benefits" to people of doing these things are less clear than the obvious benefit in time saved of speeding. I don't understand any more than you do why anyone would fail to use a seat belt; it only takes a few seconds to buckle up. (I don't ride motorcycles so I won't try to speculate on why people would not wear helmets.) People obviously derive pleasure from drinking, but that doesn't mean they necessarily derive pleasure from drunk driving; that seems more like bad planning than anything else. So it might not be quite as costly to try to regulate these behaviors as it would be to try to regulate speeding. But I think it would still be a lot more costly than airport security is.

Education is a different matter; but I'm not sure there's much that can be done with education that isn't already being done. Are there really many people who haven't been told that seat belts are a good idea?

The only other suggestion I have is to impose more severe consequences on the kinds of decisions that can significantly affect risk. For example, consider this modest proposal: if you are involved in any accident and you are found to be driving drunk, even if no one is injured or killed, you lose your driver's license, forever. Or consider this: allow auto insurance companies, and health insurance companies to the extent that their coverage extends to injuries suffered in auto accidents, to make you pay a higher deductible, or possibly even deny part or all of your coverage, if you are injured in an accident and are found not to have been wearing a seat belt, or you are found to be at fault and to have been driving at a clearly unsafe speed. (This would have to be explicitly stated in the policy, of course.) That would make people stop and think.

[seanmcdirmid]

Speed enforcement is a political issue these days, not a cost issue, it is quite cheap to put speed cameras everywhere and many countries do. We could eliminate speeding if we wanted to, Americans just don't want to. So what if a few more people die every year...freedom!

[pdonis]

The problem I have with speed enforcement as such is that speeding, in and of itself, doesn't harm anyone. Note that in my suggestions above I didn't say anything about speeding by itself; I said if you are involved in an accident and your speeding is a contributing factor, that should increase the negative consequences you suffer. That would give people an incentive to properly consider the risks of speeding; and the person who is actually driving the vehicle, and sees the road conditions and other variables as they are, is in a much better position to judge what is a safe speed than a bureaucrat writing speeding laws.

[Dylan16807]

Oh shut up with half-baked emotional arguments, if a 'few people' dying in accidents was always a good enough reason to slow down traffic there wouldn't be a road in the world faster than 20mph.

At any rate, speed limits are set extra low with the expectation that people will go faster than the number on the sign. Trying to 'eliminate speeding' by itself is a nonsense goal.

[seanmcdirmid]

I'm not making any arguments. Is it ok to go 90mph when the speed limit is 75mph? Even if its straight I90 freeway in the middle of Central Washington? I'm not sure, but today we have that choice as enforcement is choppy.

Anyways, we'll be into self driving cars in 10 or 15 years regardless, the question is almost outdated.

[Dylan16807]

"So what if a few more people die every year...freedom!" is the emotional and clearly-wrong argument I'm referring to.

As to your question, it depends on the road and 15mph is an unusually large number. I'm used to things more like '70-72 in a 65 zone' and '39 in a 35 zone'. On a straight freeway 90 is probably safe.

But my point was that the current system is broken in such a way that rigorous enforcement of the exact numbers on speed limit signs would make things even worse.