pageok
pageok
pageok
Becker & Posner on Libertarian Paternalism:

At the Becker-Posner blog,, Becker and Posner discuss "libertarian paternalism" (albeit in reverse order. Here is Judge Posner's take, and here is Becker's critique.

American Psikhushka (mail) (www):
Cross-posted on Becker's critique:

I can't get over how dangerous the "conflicted selves" reasoning tends to be. It's the modern academic version of the logic of the Inquisition. Whatever we do is justified because it rests on the assumption that we are doing it in the actual best interests of the individual. The rack, the flail, the hot iron? Trivial compared to saving the heretic's immortal soul.
1.15.2007 3:38pm
Richard A. (mail):
Posner's argument is philosophically sound but he is wrong about smokers' estimate of lung-cancer risks. This is rather strange given that the following study is representative and noncontroversial in the field:

Many smokers have unrealistic ideas about just how dangerous their habit really is. A new study by researchers at Rutgers University and the National Cancer Institute found that many smokers underestimate their risk of developing lung cancer and buy into common myths that make smoking seem less harmful.
The findings come from the Health Information National Trends Survey, a national telephone survey of more than 6,000 US adults, including 1,245 current smokers. Among other questions, the survey asked people to judge their own risk of developing lung cancer and the chances that other people, both smokers and nonsmokers, would develop lung cancer. The results were published in the journal Tobacco Control (2005;14: 55--59).

Current smokers thought their own risk of lung cancer was much lower than that of the average smoker. When asked to compare their lung cancer risk to that of a nonsmoker, about 21% of smokers said it was only a little higher, 23% said it was about twice as high, 22% said it was 5 times as high, and 23% said it was 10 or more times higher.

In reality, a smoker's lung cancer risk can be between 10 and 20 times higher than that of a nonsmoker, depending on how many cigarettes are smoked and how long the person has smoked.
1.15.2007 4:28pm
frankcross (mail):
That's one study, but Posner is probably referring to research by Kip Viscusi. Smokers were asked what is the probability that they will die from smoking and they substantially overstated the probability.
1.15.2007 4:35pm
David M. Nieporent (www):
Richard A: you're looking at relative risk. He's talking about absolute risk.

(Relative risk is a pretty useless measure, without knowing what the baseline actually is.)
1.15.2007 4:40pm
Shelby (mail):
Wouldn't "libertarian paternalism" be letting people individually and voluntarily delegate decision-making power to the government? I could ask NYC to tell me whether trans-fat foods are excessively dangerous to my health, and if so, which restaurants to avoid.

I can't distinguish Posner's endorsement of the trans-fat ban from paternalism pure and simple.
1.15.2007 5:32pm
Fub:
Shelby wrote:
I can't distinguish Posner's endorsement of the trans-fat ban from paternalism pure and simple.
I agree. It seems to be a distinction without a difference.

One real phenomenon that both Posner and Becker didn't address directly is the downright totalitarianism engaged by bureaucrats with irrational constituencies and their influence on actual legislation.

These are closely related to the inquisition analogy that American Psikhushka made above.

For a minor but illustrative example, as Californians voted to increase retail tobacco taxes and to funnel the money into a state agency chartered to provide anti-smoking education, the irrationalists and totalitarians took control. They attempted, I think successfully, to ban sales of candy cigarettes and bubble gum cigars, under the rubric that those products miseducated children.

Similar examples can be found in bans of toy guns, bans on "violent" television cartoons, bans on computer games, and bans on advocating use of drugs on radio and TV.

This is no longer rational "education". Its motivation is indistinguishable from that of any other Orwellian "thought control" regime, whether Maoist, fascist, or the Inquisition. Its rationale is the same: to save other peoples' "weaker selves".

The only hope I see is over the horizon. Perhaps in several generations, our far wiser adult great grandchildren will see substance prohibitions and their accompanying thought control regimes in the same light we now view slavery, indentured servitude, fascism and communism: odious relics promoted and practiced by the mendacious of a past generation and believed in only by fools.
1.15.2007 7:38pm
theobromophile (www):
Posner assumes that an individual would need an exorbitant amount of information to properly assess whether or not to dine at a trans-fat free restaurant. In reality, one only needs to know that they are bad and reducing them is good.

Food labeling laws require disclosure about the contents of fat (saturated, unsaturated, trans), cholesterol, sodium, and sugars in food. Most Americans could not even give a ballpark estimate of how much saturated fat (in grams) is unhealthy, nor explain how it differs from unsaturated fats, but we assume that the labels are sufficient.

Can't help agree with Shelby. Posner spends an entire paragraph discussing individual issues with determinging a healthy amount of trans fat from restaurant meals. He then somehow uses lack of uniformity between individuals to advocate for removing the individual from the equation via a government-imposed uniform standard (i.e. the ban). I don't get it.... :(
1.15.2007 7:50pm
EricH (mail):
We must force men to be free by freeing the mind of error.

The body too.

Possibly the most dangerous sentence in the history of mankind. If by "freeing" we mean (as the author* did) coercion.

Okay, so I exaggerrate only (mostly?) slightly.

[*Rousseau]
1.15.2007 10:21pm
EricH (mail):
We must force men to be free by freeing the mind of error.

The body too.

Possibly the most dangerous sentence in the history of mankind. If by "freeing" we mean (as the author* did) coercion.

Okay, so I exaggerrate only (mostly?) slightly.

[*Rousseau]
1.15.2007 10:21pm
J. F. Thomas (mail):
That's one study, but Posner is probably referring to research by Kip Viscusi. Smokers were asked what is the probability that they will die from smoking and they substantially overstated the probability.

Well, smoking will kill one in three smokers. I doubt smokers overestimate that risk. What they probably do overestimate is the risk of lung cancer while underestimating the risk of emphysema, heart disease and other smoking related diseases. In fact Posner pulls a fast one by saying


The dangers of smoking are well known; indeed, they tend to be exaggerated--including by smokers. (The increased risk of lung cancer from smoking is smaller than most people believe.)


Notice how he goes from claiming all the dangers of smoking are exaggerated to merely citing that the increased risk of lung cancer is smaller than most people believe.

He would probably be a damn good advocate for the Tobacco industry.
1.15.2007 10:41pm
frankcross (mail):
Actually, I think that is what the study found. As I recall off the top of my head, people believed that smoking was >50% likely to cause their death.
1.15.2007 10:53pm
frankcross (mail):
Should have checked before posting. You were right that the overestimate was for lung cancer. For overall mortality about 30% overestimated and 40% underestimated. But these were based on a baseline 50% probability of death from smoking. If the true probability is 33% they probably overestimated that too.
1.15.2007 10:57pm
Richard A. (mail):
In addition to being wrong on perceived risk of lung cancer from smoking, Posner is also wrong on the risks from trans fats: "and if as I believe the dangers clearly exceed any benefits from trans fats compared to substitute ingredients ..."
The most common subsitute for trans fat is good, old-fashioned saturated fat, i.e. lard. Restaurants are free to use lard, bacon grease and even butter, which has much more saturated fat than lard.
These are no better for you than trans fats. They do taste better, however, but I don't think that was Posner's point.
1.16.2007 12:07am
David M. Nieporent (www):
But these were based on a baseline 50% probability of death from smoking. If the true probability is 33% they probably overestimated that too.
They're both drastic overestimates, because they include people who'd have died anyway. When a 90 year old person dies of heart disease, it's absurd to call it a death caused by smoking, even if statistically it's true; the mortality rate of non-smokers, as well as smokers, is 100%, and if we don't die of heart disease, we die of something else. The relevant question (*) is years of life lost to smoking.


(*) Relevant for mortality purposes. There may be quality of life issues -- but that's a lot harder to measure, and as Viscusi points out, people ignore the benefits of smoking.
1.16.2007 4:32am
Andy Freeman (mail):
> In reality, one only needs to know that they are bad and reducing them is good.

Ah, to live in a world where things have benefits or costs and not both, and where only one factor matters.

Did you ever notice that folks pushing that line never seem to let other people pick that factor?
1.16.2007 8:43am
J. F. Thomas (mail):
Ah, to live in a world where things have benefits or costs and not both, and where only one factor matters.

Well, in this world there are some things that do not have many benefits, yet the costs are great. I am still racking my brain to figure out what the "benefits of smoking" are other than the nicotine high.

As for trans-fats, its true that some of the more traditional substitutes like lard and butter may be equally bad for you, albeit for different reasons, but that does not change the fact that trans-fat is itself unhealthy. And although restaurants may substitute unhealthy alternatives, there are healthier substitutes.
1.16.2007 9:42am
theobromophile (www):
Well, Mr Freeman, since I'm a hard-core libertarian, consider me a statistical error in your theory. :)

Posner's argument (I think) is that, since there is no way for an individual to properly assess the exact benefits, the government should do it. I think that people are perfectly capable of deciding if the increased cost is worth an indeterminable benefit. (Part of my opposition to the trans-fat ban is cost: upper-class people don't want icky stuff in their food, so everyone has to pay more, regardless of how much they care about said icky stuff.)
1.16.2007 1:06pm
David M. Nieporent (www):
Well, in this world there are some things that do not have many benefits, yet the costs are great. I am still racking my brain to figure out what the "benefits of smoking" are other than the nicotine high.
Well, I guess we're even, because I'm still racking my brain to figure out why "People enjoy doing it" does not constitute a benefit to you.
1.16.2007 2:16pm
J. F. Thomas (mail):
Well, I guess we're even, because I'm still racking my brain to figure out why "People enjoy doing it" does not constitute a benefit to you.

I'm not dismissing there is a benefit to smoking (obviously, a nicotine high is worth something, although the addictive effects probably overwhelm the mild euphoric benefits and become the main reason people continue to smoke). But considering the overwhelming negative costs, you can't seriously argue that there is any rational "Posnerian" explanation for smoking. Just from a economic standpoint, ignoring the health effects, the "high" you get from smoking a pack of cigarettes has got to be one of the least cost effective (legal) ones available.

I would like Posner to do an economic analysis of why people smoke. It would probably match his unintentionally hilarious economic explanation of homosexuality. In which he is able to explain male homosexuality from an economic basis but is completely befuddled by lesbianism. "There's got to be an economic explanation, but hell if I can figure out what it is, but damn is it hot!"
1.16.2007 3:20pm
David M. Nieporent (www):
Shorter JFThomas: "If people disagree with me, they're irrational."

The fact that you think the negative costs are "overwhelming" does not mean other people do. They simply weigh the costs and benefits differently than you do, just like skydivers, mountain climbers, or others who engage in risky activities that I have no interest in.
1.16.2007 4:03pm
J. F. Thomas (mail):
"If people disagree with me, they're irrational."

Remember, we're discussing Posnerworld here. Everything, literally everything, has to be broken down into a cost-benefit analysis. The adrenaline rush you get from skydiving has to be better and cost less than an injection of adrenaline. The man has an abacus for a heart. In his world smoking only makes sense if the accumulated costs of smoking are outweighed from the pleasures derived from it. It seems to me that the costs associated with smoking are awful high compared to the benefits, which even you seem to admit to extend much beyond a rather mild high.
1.16.2007 4:19pm
Mike S. (mail):
The most common subsitute for trans fat is good, old-fashioned saturated fat, i.e. lard. Restaurants are free to use lard, bacon grease and even butter, which has much more saturated fat than lard.
These are no better for you than trans fats.


There seems to be some evidence that trans fats are more dangerous than saturated fats. "Ounce for ounce, trans fats are far worse than saturated fats when it comes to heart disease. The Nurses' Health Study found that replacing only 30 calories (7 grams) of carbohydrates every day with 30 calories (4 grams) of trans fats nearly doubled the risk for heart disease.(9) Saturated fats increased risk as well, but not nearly as much." http://www.hsph.harvard.edu/nutritionsource/fats.html, citing to Willett WC, Stampfer MJ, Manson JE, et al. Intake of trans fatty acids and risk of coronary heart disease among women. Lancet 1993; 341:581-5. (While I've seen enough cycles of conventional wisdom that I don't take that as the final word, it appears to be where the evidence is pointing for the moment.)
1.16.2007 9:46pm
Richard A. (mail):
Sloppy work by Willett et al. since 4 grams of trans fats equals 36, not 30 calories (4 grams times 9 calories per gram) whereas 7 grams of carbs is actually 28 calories (7 grams times 4 calories).
I suspect that is not a direct quote from Willett, who in my experience would never make that type of mistake.
1.17.2007 12:54am