Damn Liars

Reader Mike Geary points us to the latest horror story of life in these United States:

Highway Deaths Hit 13-Year High in 2003

The number of U.S. traffic deaths rose nearly 1 percent in 2003 and reached a 13-year high at 43,220, the government reported on Wednesday.

It was the fifth straight year road deaths rose, although passenger car fatalities decreased. Sport utility vehicle deaths went up roughly 10 percent over 2002, with more than half of the victims in those crashes killed in rollovers. Motorcycle deaths also jumped.

After such a dramatic opening, it's comforting to read the punch line:

Despite the increase in the annual death count, the fatality rate per 100 million vehicle miles traveled remained constant at 1.5 deaths because more people were on the road.

Somehow a headline like "Automobile Fatality Rate Stays Constant" is just not going to grab eyeballs like one that reads "Highway Deaths Hit 13-Year High in 2003."

Comments

[Ruthless]

Ike's interstate highways have been more costly than Social Security and more deadly than WWIII.

The history of trains and US government meddling was equally crappy, but if trains had been totally government-free all these years, how many highway deaths do you think we'd be having now?

[joe]

Traffic fatalities per 100,000 miles driven had been declining every year since the 1920s. Then, in the early 1990s, the rate of decline began to decrease. A few years ago, the rate of fatalities (again, the rate, not the raw numbers) levelled off. Something happened in the US automobile market in the early 1990s that caused a 70 year trend to slow, then stop.

Can anyone think of a significant change in the American automobile market that coincided with this dramatic trend? One that auto safety experts contend has made the roads less safe?

1990s. Changes in automobile market. Linked to safety issues...hmmmmmm.........

[ted koppel]

I am going to read all the names of the fatalities - don't miss the next Nightline!

[Feqmxa]

F'n A. Isn't that what it always is?!?!?

[Jean Bart]

I have read that per miles travelled, most forms of transport have about the same fatality rates.

[joe]

What matters is PASSENGER miles travelled, since rail and bus (the safest) have many more passengers travelling each vehilcle mile.

[Jason Ligon]

joe:

I know, I know. You are talking about CAFE and similar measures kicking in so that there is less steel in every automobile except trucks , right!

Where's my prize?

[Jason Ligon]

Hmm, maybe joe could be talking about the mathematics that would more or less predict that even a 70 year trend of declining values would taper off, unless one believes that reduction to 0 from 10 is just as easy as a reduction to 990 from 1000.

[Russ D]

"Increased congestion on the roads would tend to reduce fatalities, if anything."

Where's the logic for that conclusion? Reduction in speed due to the crowding?

If a one-block stretch of road had 10 cars on it, and I spun out in that stretch, there are 10 possible driver fatalities besides my own. If the stretch is more congested, say 11 cars, there are 11 possible driver fatalities.

If highway deaths are rising, wouldn't that be a reason to consider using public transit more?

Jason doesn't mention that CAFE and the required safety features on cars are part of the cause of the rise in the SUV market since these vehicles didn't have to meet those requirements hence keeping their manufacturing costs down and their resulting sticker price down.

[jc]

Don't you want idiot drivers off the road? I know I do. If that means the idiot is off the road because of dying in a rollover, isn't that the most efficient way?

And wouldn't a higher number of taffic fatalities be an INCENTIVE to take public transit?

[JDM]

"Can anyone think of a significant change in the American automobile market that coincided with this dramatic trend?"

You're talking about the fact that our roads are more crowded than ever because urban planner types are trying to increase traffic conjestion in an attempt to to force everyone onto mass transit. Right?

If so, you're wrong to describe it as a trend in the automobile market. It's more the result of out of control technocrats using their positions to radical political agendas on clueless local politicians.

[JDM]

Errr.. The word "push" should occur somewhere in that last sentence. In keeping with quiz theme, can anyone figure out where?

[Shannon Love]

Mo,

Everything you say is true, however, this is an area where scale and degree are very important. Merely, saying something is more or less "safe" without putting concrete numbers to it tells us little. It's as useful as saying that both baseball and basketball players are taller than average. An important matter of degree gets missed.

From memory, the fatalities per million passenger miles for different vehicles runs something like: passenger vans 2.80, sports cars 2.50, two door passenger cars 2.00, SUV 1.55, small 4-door 1.48, large 4-door 1.40, minivans 0.90. It's easy to see that while SUV's are less safe than sedans they are not dramatically less safe. If everybody who drove an SUV switched to a sedan safety would not improve dramatically but if everybody who drove a sports car switched to an SUV safety would improve dramatically.

SUV get negative attention wildly out of proportion to their overall safety. The hatred for SUV's springs not from safety numbers but from morally outraged secular puritans who view SUV's as emblems of conspicuous consumption and environmental destruction. In furtherance of this moral crusade they cherry pick negative statistics about SUV and disseminate them widely while actively downplaying the positive statistics.

However you cut it, it's intellectually dishonest and a piss poor way to make public policy.

[Douglas Fletcher]

I love my pony, but he's a little shy in traffic.

[chuckR]

Jason

To your equations, you need to add 1/2mv**2 = 1/2 sigma*epsilon**2*volume. This is the conversion from kinetic energy to strain energy. Modern cars fold up like cheap umbrellas at their corners and the crumpled metal diverts crash energy into strain energy and heat. As for the crash-up, your 1975 Olds '88' or Buick deuce and a quarter against a 2004 Chrysler 300C. See the link (hmmm, may be busted..) for a discussion by Thomas Wenzel, P.E. of SUV vs. mid-size sedan safety - a lot of accidents involve a failure in the wingnut - behind the wheel - regardless car type. SUV drivers drive up to a perceived level of risk just like anyone else - that happens at a lower speed than in a car due to their high cg and suspension that is compromised by 4wd/offroading requirements.
Unfortunately, discussions of SUV safety quickly become a matter of religious conviction.

[JDM]

"SUV get negative attention wildly out of proportion to their overall safety."

Now come on. joe is a professional urban planner. He wouldn't let his extraordinarily evident political bias get in the way of providing us with the most fair overall view of the data, would he?

Also, in my previous post, the word "pushing" should replace one of the words in the post. Can anyone figure out which?

[robera]

Does it break down by state? Cuz the million people who moved to LA County last year--from places where they don't seem to have drivers' ed-- HAVE to have caused a zillion accidents. And then our former gov. gave then licenses.

[robera]

Does it break down by state? Cuz the million people who moved to LA County last year--from places where they don't seem to have drivers' ed-- HAVE to have caused a zillion accidents. And then our former gov. gave then licenses.

[robera]

Does it break down by state? Cuz the million people who moved to LA County last year--from places where they don't seem to have drivers' ed-- HAVE to have caused a zillion accidents. And then our former gov. gave then licenses.

[joe]

Sorry JDM, the fact that congestion results in lower fatality rates has been definitvely proven, and is an accepted fact in transportation engineering. I have never seen a transportation engineering textbook that drew any other conclusion. You really do like to demonstrate your ignorance, don't you?

Jason, the modern equivalent of a 75 olds is not a Ford Focus. You drive a 69 VW bug, I'll drive the Focus. Or, you drive the Olds, I'll drive an Accord. Bring it on. My car might look a little worse than yours from the outside, as I call the morgue for you.

"you might think the numbers of people being smashed by SUVs would be balanced out by the fewer deaths of the people in the SUVs, statistically speaking." Actually, they do. The increase in safety enjoyed by SUV drivers comes at the cost of those people they kill. But that increase is offset by the fact that SUV drivers are more likely to get in an accident in the first place. Overall, SUVs have the same deaths/100k miles as midsized passenger cars, lower deaths/accident, and higher accidents/100k miles.

Once again, the CAFE argument falls flat. A 2002 Malibu is safer to drive than a 1978 Malibu (which rule, btw), despite the heavier car's size, because of superior engineering. If the smaller, better mileage cars of today were less safe than those of yesteryear, death rates would be going up. In the real world, they went down for 20 years after the imposition of CAFE, through repeated increases, a decline which only stopped when SUVs became common.

As for SUVs and idiotic CAFE exemption, I'm all for getting rid of it. Get the idiotmobiles off the road, and watch the trend of declining deaths re-emerge.

[Mo]

Shannon,
I don't think these fact should be used in public policy. If more people want to drive SUVs, I don't care. Well, I do care because I think they make the roads more dangerous for me, but so do 16 year olds and I'm not going to legislate them off the roads. You can't really measure how unsafe the roads become because of decreased sight lines (i.e. if a sedan is in front of me, I can see if the car in front of them slams on their brakes and gives me extra time to respond. With an SUV in front of me, I have to rely solely on the vehicle in front of me).

BTW, you commit the same fallacy as I did. There are more SUV drivers than sports car drivers. I don't know the degree (I'd say 3:1 is a conservative estimate based on my observation), so if all sports car drivers went to SUVs and all SUV drivers drove minivans (which is what they largely replaced) roads would be safer overall. Sedans instead of minivans would give the advantage to SUVs but it would be closer. 🙂

I hate SUVs because of persoanl experience. When a vehicle tries to enter my lane while I'm still in it, 90% of the time it's an SUV because it's harder for them to see me. They make it harder for me to see and make my life more dangerous and expensive (higher insurance and gas prices). It has nothing to do with a religious belief that all consumption is bad and they damage the environment. It's all about me, me, me. But people are still welcome to drive what they want.

Actually, I would change one legislative piece regarding SUVs. I would remove the tax break for small businesses purchasing the behemoths. I wouldn't cry if they were held to the same CAFE standards as my car, whether its none or full.

In summary, my issue is less with the preferences of others, but the costs they impose on me (insurance and decreased safety).

[fredH]

Did anyone else notice the line about motorcycle deaths increasing? As someone who has been riding bikes for three decades, I would speculate that this might be due to the large number of middle-aged people who seem to be buying bikes for the first time. There seems to be a rapidly growing Cult of the Harley among aging yuppies indulging in nostalgie de la boue . Handling a motorcycle is one of those things that is best learned while young. That way by time time you're my age, you have experience to offset the slowing reflexes.

[joe]

fredH, I think the increase in congestion might be a partial explanation. More low-speed accidents occur on congested streets. These accidents are extremely unlikely to kill people in cars, but people on motorcycles are so damn vulnerable to anything.

[garym]

Russ D: Yes, it's the reduction in speed due to crowding which reduces fatalities when congestion is high. It's always struck me as counterintuitive that Massachusetts has a low traffic fatality rate, but the figures are there, and that's the most plausible explanation I've heard.

If the highway is more congested, there are potentially more fatalities, as you point out, but in absolute numbers, not in fatalities per passenger mile.

[JDM]

"The problem with measuring the relationship between accident costs and traffic volume is that traffic densities are highly correlated with time of day and degree of urbanization, both of which affect accident rates and severities strongly. For example, in Zho and Sisopiku's study, many of the hours with low volumes are in the evening or on weekends, when visibility and alcohol probably play a role in elevating accident rates. Unfortunately, researchers have not yet done a careful multivariate study of the determinates of accident rates, even though this is obviously a critical issue in measuring safety externalities."

from http://www.rhsmith.umd.edu/lbpp/tpug/urban_transportation.html

So, no definitive proof, I guess. Though it could be so. I wouldn't expect to be able to trust joe to tell us that.

[Jason Ligon]

"These accidents are extremely unlikely to kill people in cars, but people on motorcycles are so damn vulnerable to anything."

Surely we should legislate in favor of kinematically inferior vehicles ...

[Russ D]

My example was one extra car, resulting in a ten perent increase in traffic, but there was probably no effect on speed with that extra car. Perhaps it's higher speed roads with more traffic, but not necessarily enough traffic to effect a congestion-related speed reduction. I'm thinking Tri-State Tollway in Chicago, where the posted speed limit is 55 but the typical speed is 70.

[Shannon Love]

Mo,

Hah, your right, I did screw up. The proportion of vehicles in each category would strongly influence the improvements in overall safety if people hypothetically switched vehicle types in mass.

What I actually, intended to show however, was that SUV's clustered in safety with 4-door passengers cars and aren't some horrific safety outliers as they are often portrayed. As an overall design SUV's are not particularly dangerous.

[steve]

This is a bit OT but...

I watched "Runaway Jury" last night. In the movie (Gene Hackman, Dustin Hoffman, John Cusack, Rachel Weisz) the great boogey man is gun manufacturers and their callous disregard for their fellow American citizens, which, of course, manifests itself by their continuing to produce and sell their product even though the use of thier product sometimes results in innocent people being injured and killed.

Anyway, one of the statistics mentioned in the movie - which I have not corroborated - is that more than 30,000 innocent people are killed per year by guns. This is the primary justification given for suing the gun manufacturers, arguing that through callous negligence they are responsible for these deaths.

Well, with 43,000+ road deaths per year, shouldn't we be suing the car manufacturers who are knowingly producing, and profiting from, a product that results in the deaths of innocents? I mean, we have how many (50+) years of evidence?

[Shannon Love]

Joe,

I think you are missing the point in impact of CAFE on safety. What we have here is a Red-Queens-race were technology increases safety while at the same time CAFE standards decrease it. It is easy to show that lighter vehicles of any particular era are more dangerous than heavier cars. It has been shown that every 100 pounds decrease in weight within the same vehicle class. creates a statistically significant rise in fatalities. If there were no CAFE standards then improving technology would have driven down fatality rates much further.

The stagnation in safety rates actually began around 1977 when people began switching to much smaller cars. CAFE made fuel efficiency the primary design consideration by law. Before then car manufactures could design first for safety then for fuel efficiency. CAFE put a stop to that.

[garym]

Buried in that article is the interesting bit: "Runge said those figures underscored the need for states to adopt standard safety-belt enforcement laws..."

People without seat belts endanger no one but themselves, so enforcing those laws does nothing to make the highways safer. But such laws do allow governments to exercise that delightful feeling of taking away people's private spheres. As usual, distortion of statistics is a means to power. If the politicians can convince you that the guy in the next car is menacing you by not having a belt fastened, then they win.

It would be useful if the article said something about trends in passenger car and SUV usage, so that the numbers actually meant something. But that wouldn't promote the government's goal of controlling people.

[Mo]

Shannon,
Those number show how safe SUVs are for their drivers. Which, basically, shows that the people that say, "Let me have my SUV, it makes my family safer," are mistaken. This does not mean they do not make other cars safer. The reason why I have more of an issue with seat belt laws is that the person you make safer is yourself. In an SUV, you make other people less safe (granted, just getting behind the wheel makes others more unsafe than the marginal difference between a subcompact and an SUV).

So the solution is that sedan and minivan manufacturers need to better sell the safety of their vehicles. People assume that bigger and heavier is safer. I'm betting if people were shown that SUVs are marginally less safe than a sedan and far less safe than a minivan, there would be a shift in purchasing preferences. It would be nice if the same standards that applied to cars applied to SUVs. By this I mean, ether SUVs need to adhere to CAFE standards or no one should.

[Mo]

Shannon: I was half tempted to make a snarky Jean Bartesque insult to you. But I figured I can't beat a Frenchman in that department. 🙂

[Evan Williams]

Yeah, Steve, I saw that movie too. As I watched the credits, I looked for the writing creds. I couldn't help but feel that the whole thing was written by the Trial Attorneys Association of America. Either that, or Mike Moore. I felt like I was watching the second half of "Columbine", where Moore screws himself (and ruins his premise) by harrassing Charlton Heston and K-Mart execs. Ugh.

Maybe the next movie like this will be about suing the fast food restaurants for making people fat. And maybe Michael Jacobsen or Kelly Brownell will play starring roles. I can't wait!

[joe]

Bzzt, sorry Jason. CAFE standards were first introduced in the 1970s, and the smaller, lighter cars being built today are significantly more crash-worthy than the heavier gas guzzlers of the 50s, 60s and 70s.

You might want to learn something about a subject before you spout off like that. It's embarrassing.

[joe]

This flattening of the death rate has occurred during the same period that old, unsafe cars (no seatbells, non-shatterproof glass, hard interiors, non-lined gas tanks, etc) have virtually disappeared from the roads, and advanced safety devices like anti-lock brakes, airbags, third brake lights and like have become commonplace.

In other words, there are significant forces that should be pushing fatality rates down, contrary to Jason's assertions that increases in safety are becoming prohibitive to achieve. But they're not budging.

So I ask again,what major change in the vehicles on the road that would tend to make driving less safe has occurred since the early 90s?

I guess there are none so blind as those who will not see.

[Douglas Fletcher]

Yabbut, the people in the big cars are safer than they would have been in older, smaller cars -- you might think the numbers of people being smashed by SUVs would be balanced out by the fewer deaths of the people in the SUVs, statistically speaking.

I didn't see mentioned the increase (my perception, anyway) of congestion on the roads in most of the metro areas over the last 20 years. That should have something to do with the problem, shouldn't it?

[Allen Phelps]

Joe,

Popularity of SUV's?

(Not saying I necessarily believe that is the reason, just wondering if that is what Joe is thinking about.)

[Andy D.]

I say it's either the explosion of SUVs or airbags.... though I'd imagine that airbags do more good than harm.

[Douglas Fletcher]

"I guess there are none so blind as those who will not see."

You can take your sunglasses off, Nostradamus, it's dark out.

[garym]

Douglas Fletcher: Increased congestion on the roads would tend to reduce fatalities, if anything. Massachusetts, in spite of its well-deserved reputation for lunatic drivers, has a remarkably low traffic fatality rate. It just isn't easy to get moving fast enough in Boston to get yourself killed in a crash.

More information is necessary to make intelligent analyses. A jump of 10% in a year in SUV deaths is startling, and in itself doesn't say that SUV's are unsafe, since we're comparing like to like. Did SUV usage jump by 10%? Did SUV's become less safe in 2003? Did the habits of the drivers change in some significant way?

One factor which comes to my mind is that 2003 was a very bad year economically, and people may have neglected maintenance on their vehicles.

[Keith]

Joe:

Are you fishing for the relaxation of the double-nickel limit? That is, the reinstatment of 65 MPH or 75 MPH or Montana-speed outside of congested centers?

[Andy D.]

Wait, wait... cell phones?

No no, it's probably the SUVs. Because they roll over more frequently than normal cars, they cause a disproportionate number of deaths. Sure, you're better off with an SUV if you have to go straight and ram into something, but you're much less safe if you have to turn quickly.

[Mark Draughn]

Another possible reason for flattening of the fatality rate may be that most people have as much safety as they need.

For example, all other things being equal, airbags make you safer. But people may choose to change one of the other variables in the equation by driving more aggressively, secure in the knowledge that they can reach their destination sooner while protected from the increased risk of collision by the airbag.

To put it another way, people are trading away the added safety for reduced driving time or more driving enjoyment.

[Rene Gonzalez]

Those car-driving Rambos got exactly what was coming to them.

[James Merritt]

Maybe someone should convince the Onion to branch out and establish a REAL newspaper that poses current news stories in terms of their underlying truth and impact on the everyday person, from the headline on down. In this newspaper or on this website, the headline would be "Auto fatality rate stays constant." Or, "Presidential Candidates Debate Ways to Ignore, Divert Attention from Social Security Collapse." Or, "Most Americans Unaffected by Federal Shutdown." Or "Mission Far From 'Accomplished,'" and so forth. It wouldn't be satire, exactly. It would be the truth that sounds like satire, in comparision with all the spin and hype that we normally get from the media.

[Shannon Love]

Joe,

You argue from a mistaken premise. SUV's are no more dangerous per passenger mile than comparable 4-door sedans. You neglect to consider that each vehicle design has it's own strengths and weaknesses in regard to safety. Quite often strengths and weaknesses cancel each other out. Everything is a tradeoff.

SUV's are far more prone to roll over than other vehicles except large passenger vans but they are far less likely to suffer other types such as skidding caused by bad road conditions. SUV tend to be safer for all types of accidents except roll overs. Statistically, lives saved in one type of crash cancel out lives lost in other types.

The best explanations for the stagnation in drop of deaths per mile are CAFE standards and moral risk.

CAFE require manufactures to lower the average fuel consumption of their entire fleet. The easiest way to accomplish this is to sell highly fuel efficient but relatively unsafe cars to poor people and fuel inefficient but relatively safe cars to rich people and then average the fuel efficiencies together. While contemporary small cars are more safe than large cars of the past they are not more safe than the cars that would be sold today if CAFE standards did not exist. Consequently, the drop in deaths rates leveled off.

Moral risk is caused by the increase in safety technology that causes drivers to view risky behavior as less dangerous. For example, trying to improve the safety of stretch of road by making it wider and better banked often fails because people just drive faster on the seemingly safer road. Airbags and other technology appears to have a similar effects on drivers perception of risk in contemporary automobiles.

The SUV phenomenon was created in the first place by government interference in the market. CAFE and pollution standards excepted "utility" and commercial designs creating a powerful economic incentive for people to buy such vehicles instead of comparable cars.

These types of attempts to engineer society always backfire.

[Patrick]

That reduction from 10 towards 0 might happen, if virtually unenforceable laws against such things as running stop signs were somehow taken a little more seriously. I live in a part of New York City (Queens) where drivers gleefully flip me the bird because I am inconveniencing them by lawfully crossing a street in a crosswalk (this after they *speed up* upon seeing me, and miss killing me by inches). I think a little more education is in order, in order to bring those 43,220 deaths (per year!) down. Not to say that all those deaths are caused by the sort of asshole drivers that predominate in my town, but it is clear to me from my travels that Americans are woefully poor drivers.

[Jason Ligon]

CAFE was introduced in the 70s, but the standards have not remained the same since the 70s. I thought it was obvious that I was talking about the progression of the standard beyond what a steel reinforced car could sustain.

Lets have a crash up derby between joe in an '04 Focus and me in a '75 Olds. Joe should feel very comfortable in the knowledge that his car is more crashworthy. How about the rest of you?

Embarrassingly ignorant I may be, but I do know that mv1 = mv2 and that F = d/dt(mv).

[Ayn_Randian]

Maybe if we hadn't insisted on building a massive highway system on the backs of the citizens without market forces making them necessary, alternative trasportation or "something else" could have arisen and made dependency on oil and automobil deaths disappear. The more government builds roads, the more people will use them.

[Mo]

Shannon: An SUV may be safer in a collision than most cars, but it also is far more likely to get into an accident than a sedan or minivan. SUVs have less manuverability and are significantly worse on accident avoidance. The only cars worse on injuries and deaths per mile driven are sports cars and subcompacts. Minivans, medium and large sedans all perform better and are safer because of of superior engineering and handling. SUVs are the safest per non-rollover accident, but they get into more accidents. Granted, there might be some selection bias because it seems idiot drivers disproportionately pick SUVs (I am not saying all SUV drivers are idiots).

[joe]

Shannon, no question, a heavier car is safer than a lighter car ALL ELSE BEING EQUAL. But all else is seldom equal. But when judging the health and safety impacts of CAFE standards, it would be deeply dishonest to ignore the imporvements in air quality that have resulted, and their resultant impact on human health, including deaths. In ignoring this factor, it is you who is focusing entirely on one concern, to the exclusion of others - the very charge you level at CAFE standards.

Once again, while we are talking about the impact on CAFE standards on highway fatalities, the decades-long reduction in fatality came to a screeching halt when large numbers of vehicles that don't meet those standards took to the roads. In the face of this fact, your assertion that allowing greater numbers of such vehicles on the roads would result in fewer fatalities is flawed.

[garym]

The escape clause from CAFE for SUV's, based on their technically being "trucks," is one of the reasons for their popularity. Manufacturers have to keep sales of their heavier, gas-consuming passenger cars down in order to stay ahead in the CAFE race. It makes sense for manufacturers to promote SUV's and downplay wagons, even when they make both available.

If SUV's are in fact less safe than station wagons (something which may well be true, but which I haven't investigated enough to decide on personally), then the relatively larger number of fatalities due to a shift to SUV's is itself a consequence of CAFE.

[joe]

Yes, loopholes suck.