What is Seen and Not seen
My session of verbal grappling last week led me to recall an economist whose work should be acknowledged. Sam Pelztman is a professor at the University of Chicago’s Booth School of Business. In the 1970’s he wrote a shocking paper on the nature of seatbelt laws and their impact on the rate of car fatalities. The National Highway Traffic Safety Administration reasoned that seat belts had a net gain to society of 20% per annum. Professor Peltzman took another view that this conclusion was incomplete. His argument is simple but requires an eccentric mode of thinking. Imagine the way you drive as a good you wish to purchase. Every method you choose has some cost. The slow and steady driver pays with time and the occasional bout of obnoxious horn pounding. The speedy driver will arrive more quickly, but he will increase his chances of getting into an accident. Economists love to start with the most obvious point, so please accept my apologies for doing so. The less obvious point is what happens when you mandate that everyone has to have a seatbelt, not only have, but wear. This has the effect of making reckless behavior cheaper. Cheaper in the sense of lowering the probability of injury given a particular set of behaviors. When Professor Peltzman ran the numbers and took account the presence of the offsetting behavior resulting from more people wanting to drive recklessly, he found that the net effect of seatbelt laws is zero.
Is it right to mandate that people do something that is a good idea? There are always effects that may offset the supposed benefits of a policy. The EPA is created to deal with environmental issues but the parties most directly affected, regulated companies, have cause for lobbying. The regulating organizations can become focused not on the commonweal, but their own power and survival. If taxes are increased, the benefit of evading those taxes rises as well. There is no such thing as a free lunch. There are only those lunches we are willing to pay for.
June 1, 2009 at 8:51 am
“If taxes are increased, the benefit of evading those taxes rises as well.”
Isn’t that why we (should, not really do anymore) punish tax evaders? The reason why we haven’t is because the previous administration gutted the SEC.
http://www.nytimes.com/2009/06/01/opinion/01krugman.html?_r=1&ref=opinion
About the effects of Reagan’s deregulation policies. I know it’s Paul Krugman, but you can’t argue (or at least it gets a lot harder) with a Nobel Prize in economics (and his specialty is depression-era economics).
June 1, 2009 at 2:27 pm
i think your style of driving has more to do with your individual temperament and the circumstances at hand: if i’m late, i’ll drive faster whether or not i’m wearing a seat belt.
plus there are other constricting factors for not driving recklessly, like auto insurance rates going up. why does he suppose seat belts make more people drive recklessly? has that been shown? and if it has, on what kind of sample bc i’d think the impact of seat belt laws on behavior might differ depending on how you were raised (with always buckling up or not). if it’s already habit, it might not change your driving style as much as if you’re constantly, consciously having to remember to buckle up.
June 1, 2009 at 5:55 pm
It is easy to argue with Paul Krugman since a Nobel Prize does not entitle him to make the sort of errors that are to be found in his more recent work. Reading the article you cite, I am convinced that his continued publication is dissengenuous to his audience. The SEC, an agency dedicated to securities fraud, did have inadequate resources to detect the malpractices of Madoff. All one need have done to detect his wrong doings was to talk to him personally, ask the exchange if his claims about trade volumes were true, look at his published rates of return and realize that no one ever has that consistent of return rate every year, or attempt to duplicate his methods in a simulation. Any one of these acts would have revealed that something was wrong. I cannot speak to why these things were not done by the SEC, but none of these acts would have cost that much money or time.
Going back to my objections to Paul Krugman, a cloud of non-specific statements showing his dislike for Regan is not a sound argument. There are a few interviews on the link I am providing that give better explanations of the possible reasons for the current situation. Some are more in favor of increased regulation than others, but the authors are not taking the disingenuous rout that Krugman takes to prove himself correct.
http://www.econtalk.org/archives/great_depressio/
June 1, 2009 at 6:22 pm
S: As I said earlier, seat belts make driver more reclessly cheaper. It is not an effect that we need be fully concious of for it to exist. If there is a slight diffrence in how one ear performs, you might cock your head slightly. Your manner of walking is a complex series of motions that have to take into account a number of factors, if you can honestly say that you conciously consider every single one of those motions, then hats off to you.
Peltzaman does not argue that seatbelts do nto make you safer, but he is arguing that there is an offseting effect from decreasing the risk of reckless driving.
June 2, 2009 at 8:08 am
i think you misunderstand my questions. i find it rather intuitive that the safer you feel, the more reckless you’re apt to act, though Pelztman’s application of this intuition is new and interesting to me. i was simply curious as to whether you know the breakdown of the results. it could be likely that the two groups (always worn seat belts vs now changing their behavior to wear seat belts) reacted differently to seat belt laws, and that one group might have been responsible for skewing the results to the zero net effect. it’d also be interesting to know if this effect petered off in time.
to take one of your examples, “If there is a slight difference in how one ear performs, you might cock your head slightly. ” i’m suggesting that if you know that your hearing in one ear is worse, you may overcompensate by turning your head at a greater degree than if you were unconscious of the fact.
have you come across any of his raw data? after reading a bit, it seems that a lot of the offset stemmed from the increased deaths of pedestrians, bicyclists and motorcyclists… that’s definitely an interesting result.
June 2, 2009 at 4:30 pm
The paper citation is: “The Effects of Automobile Safety Regulation.” The Journal of Political Economy, 1975, 83(4), pp. 677-726. You should be able to find it on JSTORE.
Professors Peltzman’s data is from the National Highway Traffic Safety Administration’s figures. He also pulled in data from a number of other souces which I cannot list at the moment. The model does not account for insurance rates or one or other aditonal factors that one can think of. But no model can be complete and comprehensible at the same time. To give a complete model of the physics of a pool shot would require more variables than we can solve given the current limits of mathematics. But we do not need all of the variables to get a result that will be sufficent for scientific argument. This is what astronomers haqve to do when faced with the problem of copying the music of the spheres in mathematical notes.
June 3, 2009 at 5:01 pm
I am somewhat baffled by Mr. Young’s perceived opposition to more comprehensive models. It is certainly true that adding more variables to an econometric model will increase the degrees of freedom, and therefore generate larger standard errors for estimate coefficients, but that is precisely the point. To overly simplify a model ignores many facets and factors which may or may not play a pivotal role. Having created an econometric model with over 20 independent variables, I can tell you that many of them simply drop out as they are not statistically significant – yet I would have been amiss in not including them. I am sure Mr. Young is aware of omitted variable bias (the tendency for the effects of omitted variables to be attributed to included variables). Larger models may be more messy, but they also more accurately represent reality. As the famous quip goes, “There is no free lunch” – what we sacrifice in precision we gain in completeness.
June 4, 2009 at 1:29 am
A very good point. I must admit that when I posted my previous comment, I made an error ( thats what I get for relying on memory when the origenal paper is on antoher computer). Looking back at the paper, I forgot thta insurance was taken into account as part of the costs of an accedent. I would recomend reading the origenal paper to gain a full view of the research done in this area.
Adressing the Mr. Draine’s about econometric modeling, I do not oppose complex modling. I was merely trying to point out that a model need not account for every detail to give a result useful for perdictive results. Milton Friedman once made the argument that an when one has to make a choice between perdictive accuracy and completness, accuracy may be more disierable. Besides every model has known knowns, known unknowns, and unkown unknowns. The degree to whichthe results of any model are affected by a lack of information can be determined by a sensitivity analysis, which I still need to learn how to do.
June 4, 2009 at 10:19 am
“…every model has known knowns, known unknowns, and unkown unknowns.”
You sound like Donald Rumsfeld hahahaha.
June 4, 2009 at 3:21 pm
Yes, a very poignant man at times.
June 4, 2009 at 6:18 pm
That little bit of poetic drivel from him was used as an example of “found poetry” in my creative writing class. For those who have been fortunate enough to forget:
‘There are known knowns. These are things we know that we know. There are known unknowns. That is to say, there are things that we know we don’t know. But there are also unknown unknowns. There are things we don’t know we don’t know.”
As much as I despise the man and think he should be in prison for war crimes, he is quite crafty with words!
June 4, 2009 at 6:24 pm
“Milton Friedman once made the argument that an when one has to make a choice between perdictive accuracy and completness, accuracy may be more disierable.”
But if you’re not accounting for as many variables as possible, doesn’t you “accurate” result become unusably narrow since it can’t be applied to a more realistic situation? Then again, economics is by far not an exact science – economics assumes people action in rational self-interest and we all know this is almost never the case in one way or another, be it neglect, cheating, or simply not knowing what is in one’s own best interests (or in the case of a good number of social conservatives, voting against their economic interests out of fear that the gays, socialists and abortionists are coming to get them in their sleep). One of the few things I got out of AP Microeconomics was “you can’t graph people.”
June 4, 2009 at 6:26 pm
Any model used in economics is inherently problematic since one must attempt to strike a balance between completeness and accuracy, and people choose which is more important to them.
June 5, 2009 at 1:14 am
Such actions are rational. If you know that you will not be caught for illegaly parking, the value of doing so increases. I would also ask what you mean by not knowing what is in your intrest. I would also ask how exactly do social conservatives vote against their interests.
June 5, 2009 at 2:39 am
“I would also ask what you mean by not knowing what is in your intrest.”
For example, not knowing to take out a loan on a new house when you barely have enough money to pay the principle, or falling victim to one of these predatory lenders that became legal after Congress repealed Glass-Stegal in 2002 (I think).
“I would also ask how exactly do social conservatives vote against their interests.”
Voting for candidates who are against abortion but promote economic policies that actually hurt said people. They scream about lower taxes but that really means lower taxes for the rich, not necessarily those who are duped into supporting them.
June 5, 2009 at 2:44 am
A good book regarding this is “What’s the Matter With Kansas?” Unfortunately my internet is failing and I can’t google the author. It’s a case study of Kansas about how the far right came to power by scaring the average American with words like terrorism, traditional values, and socialism, written by an ex-conservative.
June 5, 2009 at 2:25 pm
Actualy, one can make the case that lower taxes on the rich will make everyone better off. More moeny for investing and so forth.
On the nature of predatory lending, I would be carful with that catchall term. If you are speaking of those who deliberatly mislead or lie to those they lend to, then that is improper. If you are speaking about those who happend to charge high intrest rates to those who have poor credit, then that is not wrong. Unpleassant yes, but not wrong. Intrest rates are a way of compensating fo the lender taking on the risk of lending money. So higher risk should mean higher rates, otherwise the worng signals would be sent out and we would have an glut of defaulting loans.
I cannot speak to the multitude of resons why one would take such a loan, but there are a numer of cases where such loans would be in their intrest. For example, if the housing maket is on the rise, then a high intrest loan is not a problem since you can always sell the house for a net gain.
The Glass-Stegal act that you may be referring to was only partialy repealed and that repeal allowed investment banks and commercial banks to combine with each other. This allowed banks to have the stability and capital that an investment bank has with a diversified portfolio and thus alowed better service on the comercial side.
June 5, 2009 at 2:52 pm
“Actualy, one can make the case that lower taxes on the rich will make everyone better off. More moeny for investing and so forth.”
We’ve seen what happens with that – the 20s, 80s, and the past 8 years. When will we realize that trickle-down economics doesn’t work?
“If you are speaking about those who happend to charge high intrest rates to those who have poor credit, then that is not wrong”
The whole point of regulation is to prevent banks from doing that. The banks got reckless with who they lent money to and then AIG insured all of those loans, hence when no one could pay their massive-interest loans, the house of cards collapsed.
“This allowed banks to have the stability and capital that an investment bank has with a diversified portfolio and thus alowed better service on the comercial side.”
Uh, what stability are you speaking of? Last time I checked they’re one of the main reasons for this mess.
June 6, 2009 at 9:48 am
Can you speak to specific instances in these decades you mention?
On your second point: Are you saying that the point of regulation is to prevent price signals from working correctly, because that is what happens when you prevent intrest rates from rising. If gas prices were not allowed to rise, then people would over consume when gasoline is scarce and the lines of the 1970′s return. If I introduce more price controls, that begats more misallocation of resources untill one is faced with the truth that prices cannot be controlled by a central authority. This is what happend in West Germany following WWII and the USSR near the end of the Cold War.
“Uh, what stability are you speaking of? Last time I checked they’re one of the main reasons for this mess.”
Not all investment banks are to blame and those that purchsed these securities had an idea that backfired. That is what happens when men inovate, sometimes those inovations fail. The instituonal memory should prevent this particular set of circumsances from happening again, but some other finacial inovation will come along that may also fail. Recessions are not a pleasant thing, but they are not preventable. When it is time for a forrest to burn, it will brun no matter what you do to it. It needs to burn inorder to continue. That is what a recession is, a burning away of excess investment in one area so the market may reallocate to a better venture.
Banks are better provides now than they ever were. Do you remember why banks used to give out toasters with every account? Because regualtiors insisted that banks could not compete on intrest rates, locations, or any other service that would benifit the consumer. So they used toasters.
February 3, 2010 at 10:08 pm
the last quarter of 2009 seems promising as we have seen lots of signs of econic recovery against the massive economic recession. i hope that in 2010 all our economies would be back on track. recession really sucks.