« Comment on the New York Ban on Trans Fats--BECKER | Main | Drunk Driving--Posner's Comment »



Feed You can follow this conversation by subscribing to the comment feed for this post.

Gabriel M.

Two wrongs make a right?
Because the US nationalized its health care insurance industry, then *whatever* Americans do, that will jeopardize their health in any way, is an externality and subject to summary regulation?

As for the myth of the knowledgeable, benevolent "expert" fit and willing to replace the self-interest, prudence and preferences of individuals... straight out of '50s Soviet propaganda. -- Why don't we take this further and have panels of experts decide on the desirability and each and every product?
(You chide your detractors for not specifying a limit to freedom. But how come you don't specify a limit to regulation?)

The Emperor

I still don't quite see the logic here. Should we ban cigarettes, too? It seems like there is a better case for that than for banning trans fats.

David Welker

"Some of the comments reflect a (natural) misunderstanding of the concept of "value of life," pointing out correctly that people do not sell their lives for the calculated value. All value of life means is this: if the average person would demand $7,000 to assume a .001 (one in a thousand) risk of immediate death, there would be a net increase in social welfare if the risk could be averted at a cost of less than $7,000."

Posner is framing this idea in its least objectionable form. Of course, I agree that social welfare is increased if we spend $7000 or less to avert a .001 risk of death. (Assuming that eliminating this risk does not also eliminate a fundamental freedom being exercised by risk taker.)

But would social welfare be increased if we spent $8000 on eliminating this .001 risk instead? That is, would it be worth spending $8 million per life saved? Many of those who advocate this sort of cost-benefit analysis would say no. It is worth spending up to $7000 to eliminate a .001 risk, not more than that. If we spend more than that, we will decrease social welfare.

Framed in this light, the objectionable nature of what is being said becomes evident. First of all, the $7 million figure is arbitrary. Not because it is not based on anything. It is, after all, derived from observing people's behavior with respect to risk-taking. But rather, because it is based on inferences from observed risk-taking behavior that are insupportable.

Let us say that someone would be willing to take a job, say washing the outside windows on sky scrapers, that increases their risk of death by .001 over the course of one year. Since there are less employees available for such jobs (a lot of people are not willing to clean the glass on high buildings, for example) wages for the job increase. Let us say by $7000. Can we move from this to saying that the "statistical value" of life is $7 million? ($7000 / .001 = $7 million).

I don't think so. Because people who wash windows on skyscrapers are not rational and either are those who don't wash such windows. Some people are scared of heights. Other people really enjoy them. Our preferences with respect to heights are not rational, and have nothing to do with the statistical risk of death. Indeed, neither the person who takes the skyscraper window cleaning job or the person who does not is likely aware that the risk of death is .001. There behavior is not likely to change when it turns out the risk is only .0008 instead of .0010. (By the way, this small change in a small risk has big implications for the value of human life. Instead of $7 million, life would be $8.75 million if the risk of .0008 were compensated with a $7,000 increase in salary.) The $7,000 increase in salary is not compensation for risk of death as rationally contemplated by the parties, but rather a reflection of those who leave the market due to their fear of heights. We cannot really say anything about how much the window washer values their own lives when the window washer and those who exited the market were not even aware of the magnitude of the risk they are taking.

Even if the window washer is aware of the risk, they do not respond to it rationally. If a 1 in a 1000 risk of death is worth $7000, then it should follow that a 990 in a 1000 risk (a 99% risk) is worth 990 * $7000 = $6.93 million dollars. But if we were to ask window washers if they are willing to take a 99% risk of death for a mere $6.93 million dollars, the vast majority (probably all those without severe emotinal problems) would say no. This indicates that the window washers were not being rational with respect to the .001 risk of death that existed before. But of course, if people are not acting rationally with respect to the .001 risk, then their behavior with respect to that risk is not a basis for arriving at a monetary value for human life.

This should not suprise us. People are hardly rational, in any meaningful sense of the term, when it comes to the prospect of their own death.

Of course, I don't even have to resort to the point that people are not rational to make my point. (Even though this is a very strong argument in this context.) I could just point out that someone working on a skyscraper may actually get non-monetary compensation from the risk itself. Being very high above the ground, with its accompanying very small risk of death, is a thrill to some people. Indeed, people are willing to pay to take what they perceive as small risks. Hence the popularity of things like sky-diving. It should be noted that this thrill comes from taking the small risk .001 and doesn't increase as the risk becomes bigger. (Indeed, likely just the opposite. Sky diving is fun when one thinks they will probably walk away from the experience. But if the risk of death was 50%, what before was fun now is likely to be tramatizing. The important point here is that larger risks don't translate into more utility. Which implies that risk must be compensated at higher rates per unit of risk as it moves higher.)

Let us say that the value of the thrill to the person working on the skyscraper is $10,000 over the course of a year. It turns out that for taking this .001 risk of death, that the true compensation is worth $17,000. $10,000 in non-monetary experience value, and $7,000 in cash. Then the "real" value of life is actually $17 million, not $7 million.

But of course, we really don't know how much the thrill from taking the small risk is worth. I pulled the $10,000 out of my hat. But maybe it really is $20,000. Making the "real" value of life $27 million, not $7 million. Or maybe its only $5,000. Making the value of life $12 million. Actually, maybe more people are traumatized by feeling as though they have to work at extra risky jobs than others to make ends meet. And on average, they would pay $2,000 to avoid this trauma. (Which is seperate from the compensation needed for the risk of death itself.) So, their $7,000 compensation is only worth $5,000 to them. This means the value of life is really $5 million, not $7 million.

But of course, we don't really an objective way to assign a value to this non-monetary compensation, do we? So what do we do. Ignore it. Pretend it doesn't exist. Even though, as shown above, it can dramatically effect our results. We wouldn't want to have this get in the way of our dream of "objective" cost-benefit analysis that makes the difficult decisions really easy. The idea of "objective" cost-benefit analysis appeals to those who would like to think of themselves as "thinking with their head instead of with their heart." But really, warping a problem and ignoring essential moral aspects is not itself a rational process. It is actually driven by one's need for certainty, which provides a kind of intellectual security. The truly rational approach is just to admit that not every problem has easy answers.

Obviously, as the libertarians commenting on this blog have observed, this sort of cost-benefit analysis can be used as a sword against their point of view. Which I am fine with, since I am not particularly sympathetic towards the libertarian viewpoint on economics. Not only that, it can act as a sword without putting regulation at significant peril. After all, at the very least, human life is worth $7 million. This is an underestimate of the benefits of the regulation. (Indeed, from my point of view, not only is it an underestimate, its a complete absurd underestimate, because you cannot attach a monetary value to a person's life.)

But for the sake of argument, let us assume life is worth at least $7 million. Obviously, it is worth more than that. How much more, we don't know. But lets underestimate the benefits of the regulation; if our analysis suggests that even these underestimated benefits are greater than the costs, this is a good argument for the regulation. Libertarian's don't win if the regulation is not actually justified by attaching $7 million to each life saved, because we can always argue that $7 million is an underestimate of the value of life, and also mention that it will prevent suffering which is not included. In this way, we are not inevitably committed to rejecting the regulation if it costs $8 million per life saved instead of $7 million.

Why not use this argument. After all, I really do believe that life is worth at least $7 million. But on the other hand, I think the risks for rhetorical confusion are too great. People will get used to the idea that a monetary value can be put onto human life, leading them to inappropriately suggest that life is $X, instead of at least $X. The effects will confusion, with people thinking that this sort of argument implies that a monetary value can be attached to human life. So, despite the attactiveness of this sort of argument, I must reject it.

Better to stick to the much more clear point. Human life cannot be monetized.

Some people like the simplicity of the sort of argument that Posner is making here. It has its appeals, given it's facade of objectivity. But objectivity is an illusion -- there is no good principled way to attach a monetary value to human life.

David Welker

Gabriel writes criticizing Posner:

"(You chide your detractors for not specifying a limit to freedom. But how come you don't specify a limit to regulation?)

Nice dodge. Well, nearly. Except Posner has specified a limit to regulation. When the costs of a regulation exceed the benefits, don't do it. This is obvious from his analysis. Cost benefit analysis has two sides. One where the benefits exceed the costs (do it) and one where the costs exceed the benefits (don't do it).

Since Posner's limitation on regulation is clear, you have no excuse for not addressing Posner's question. Are you against all safety regulations of food and drug safety? Should inspection of restaurants by health inspectors be done away with? Based on what principle?

It is interesting. It is often the case that one dodges a hard question by the claim that someone else is dodging a hard question as well. But in this case, you are incorrect. Posner did not dodge the question about limits of regulation. It is right there in front of you in the form of cost-benefit analysis.

Chairman Mao

Judge P,

Should alcohol also be banned? My concern is that government is not necessarily an impartial arbiter. Too often, special interests manipulate policy with bizarre results. Once we give the government the power to ban foods for their long-term health effects, there is no guarantee that rational behavior will prevail. Stem cell research is an example of how uncomfortable government interference can become.


I think we're focusing too much on the "value" of life, even a statistical one. I don't think anyone is arguing for the monetization of human life here. But the fact is that people themselves implicitly place a value on their life, by the amount they spend on precautions that reduce risk of death. As Judge Posner pointed out, the implied value of life in this situation is $7 million. When government, which presumably acts on our behalf when it is more efficient for it to do so rather than for individuals to act, takes such precautions, should it not act in a way that we ourselves would act? That is, if individuals take precautions with a certain implied VLS, why should government act differently?

More importantly, the question shouldn't be what the implied VLS of this particular regulation is. The actual question should always be whether more lives, even statistical ones, can be saved for the same cost. If there are other regulations out there that could save lives more efficiently, that's where our resources should go first. It's useless to consider any regulation in isolation, even with an anchor figure of $7 million/life. More important is whether the costs of the regulation could have saved more lives in a different context.

David Welker

Haris writes:

"I don't think anyone is arguing for the monetization of human life here. But the fact is that people themselves implicitly place a value on their life, by the amount they spend on precautions that reduce risk of death. As Judge Posner pointed out, the implied value of life in this situation is $7 million."

This makes no sense. First, you say that life is not being monetized, then you say the "implied value of life is $7 million." Let me see...

$7 million. That is money. Then you say the "value of life" is "implicitly" equivalent to this amount of money. That is, you are arguing for the monetization of life.

Let us see. "But the fact is that people themselves implicitly place a value on their life."

Well, my criticism is that people are not even cognizant of the "value" that you say that they "themselves" put on their life. I am sorry, this doesn't really make sense at all. Because it is not people putting a value on their life. It is the data interpreter assigning a value. Based on very questionable inferences.

Imagine a person buys a new car and spends $1500 on a feature that will decrease his probability of death by .0001 (1 in 10,000). Is this person now "implicitly" communicating that the value they place on their life is at least $15 million.

Now, if on the same day person x drives to the store to get some popcorn that they value at $5.00, and it turns out that the gas and time and price of popcorn add up to $4.00 so that there is $1.00 of consumer surplus and it turns out that the ex ante probability of death is .000001 (1 in 1,000,000) because of unusual weather conditions. We can conclude that this person only values their life at less than $1 million.

Wow. This person must be crazy or incredibly unstable. On the very same day they indicate that they think their life is worth more than $15 million, they take a risk that would lead us to believe that they value their life at less than $1 million.

Or maybe the person that is crazy is the guy interpreting this person's behavior, trying to tell a story about how this person is "implicitly" valuing their life based on the very small risks they take. The problem is that this story told is a little bit crazy. Especially when the true story is that she paid for the $1500 safety feature because the salesman suggested it and it was already installed on the car in stock. She didn't make her decision with any knowledge of statistics. And she went to the store because she wanted popcorn, not with any knowledge of the true risks involved. It never once crossed her mind that she might get in an accident.

See, the problem with your story is that it has nothing to do with the true intentions of the consumer. People are systematically unaware of the magnitude of the risks they take, especially when the risks are small. Thus, the inference that they are "implicitly" putting a "value" on their life is unsound. It is not them placing a value on their own life. It is the outsider who decides to thrust a particular arbitrary interpretation on their actions that is placing that value.

"That is, if individuals take precautions with a certain implied VLS, why should government act differently?"

Nice question. Except that it implicitly accepts the legitimacy of VLS, which is the very point I wish to bring into question. You are begging the question.

Here are a couple of points to chew on. First, VLS is not about individuals. It is about averages. No individual gets to consent to the projection of a value of $7 million by outsiders onto their life. Second, the government often does not accept individual decision-making when it comes to valuing life. (And wouldn't under your one size fits all value of $7 million.) That is why we have seat belt laws. Helmet laws. And laws against suicide for individuals who place a negative value on their life. Ultimately, the countervailing considerations when it comes to the protection of life should be other lives of others or individual autonomy, not money. That is my philosophical position, which rejects the monetization of life.

"More important is whether the costs of the regulation could have saved more lives in a different context."

This is a good point, except I think it is important to spell out the alternative regulation that will save even more lives that you are suggesting. You cannot criticize Posner's position with nothing. You need an alternative yourself.


Alex: You may not be the only one here to, apparently, underestimate the importance of restaurants and fast food flingers to the productivity and lifestyles of Americans.

"Washington, DC - With total sales expected to reach $399 billion in 2001, or four percent of the United States gross domestic product (GDP), the restaurant industry will continue its reign as the cornerstone of the nation’s economy, according to the National Restaurant Association’s 2001 Restaurant Industry Forecast. According to the Forecast, 2001 will mark the restaurant-and-foodservice industry’s 10th consecutive year of real sales growth."

""Americans demand for convenience, choice and quality food items are hyper-fueling the restaurant industry," said Steven C. Anderson, president and chief executive officer of the National Restaurant Association. "In 2001, restaurants will continue to become a more important and essential component of consumers’ lifestyles, with Americans spending more than $1 billion a day in restaurants." :

Assuming a median check of $10 or so this is 100 million "dine outs" per day and perhaps close to one dine out per wage earner, with the hefty doses of transfat heavily skewed toward the cheapest and most numerous "dine-outs".

As for "in-home" my guess is that lots of folks eat healthier at home (in regard to transfats) as they do, knowledgeably, select their oils for frying, rarely deep fry, and those who do bake at home opt to use butter and other shortenings of which they've had a chance to read the label when it came into the home. BTW the FDA was going to "expedite" the addition of transfats to food labelling in 2002, and my understanding is that it's coming on line today. I'll predict it will be a combo of legislation and consumer info that will dramatically decrease the amount of transfats in products on the supermarkets shelves... soon!

Mao: Indeed democracy is a messy process with less than impartial results and should we expect better with religious interests pitted against science (stem cells and condom or clean needles distribution to prevent the spread of AIDs.... anyone claiming rational risk/benefit policies there?) and as we have on this issue corporate and commercial interests pitted against the interests of the consumer.

On alcohol? wouldn't you credit government (us) with getting it about right? With a message of drink if you like, information on Fetal Alchohol Syndrome, raising the legal drinking age, and harsh penalties for trying to drive while impaired?

Welker: Ha-ha! Are you implying that we humans are not primarily little risk/reward calculators? And if so, getting it wrong much of the time? Being nervous about an airline trip while thinking nothing of leaving seat belts undone in the car? New Yorkers coming to Alaska grossly over-estimate their odds of being eaten by a bear, while Alaskans often over estimate the odds of getting mugged or shot in NY. (A city with lower odds of gun deaths than Anchorage) Hey, David how do you think a handheld risk calculator would sell?

"More important is whether the costs of the regulation could have saved more lives in a different context."

jjjjj: While drowning in health care costs, wouldn't it be great if we could have this as an OPEN national debate? Even if we didn't adopt most of its conclusions?

For example I'd guess it would be wise for a health care provider to offer incentives for preventative care. But what would the "nanny state" set think of providing women who took their mammogram with a voucher for lunch that might incent her to call a friend? if that's what the figures indicated? Or, as I'm SURE is the case; funding schools in our "left-behind" and largely race-based ghettos EQUITABLY and ADEQUATELY with suburban schools? Addressing the facts of our prisons being filled with those who are 75% illiterate and actually responding to the data?

Are "we" sure we favor rational risk/benefit policies? Jack

William Smith

Judge Posner,

Excellent response with good analysis. However, I think that both you and Becker did miss one crucial point about the benefits of the trans-fat ban: hosptial and Emergency Room (ER) overcrowding.
In the health care industry, of which I am familiar by practice and work, overcrowding of facilities is a major issue, especially ERs. By reducing heart disease "events", the trans-fat ban will better enable hospitals to deal with patients unaffected by trans-fat as well as reduce costs because ER visits are typically the most expensive service hospitals routinely provide.



Should we deionize our salt, unpasturize our milk, remove the fluoride from our water, unfortify the flour, declassify beef, etc., etc., etc.,? Enough!

Ray G

“if I am correct. . . if I am further correct. . . the cost . . will be shifted in part through programs like Medicaid and Medicare to the as it were nonconsenting taxpayer, then the ban is Millian.”

Lots of ifs to base an existing/pending ban upon.

More to the principle of the matter; my main resistance to so many libertarian proposals is precisely the same reason that Judge Posner gives, in that our existing nanny state necessarily would make what seem to be individual choices into various kinds of externalities.

That makes for a good argument against legalizing prostitution, for example, but it is a very poor argument in instituting new laws against already lawful practices.

In a sense, it is giving in to the nanny state and allowing the paternalism of our modern system to form all future considerations of liberty.

Whereas, opposing the legalization of currently illegal and personally harmful behavior on the grounds of externalities, is simply saying “Hey, lets fix our system first, then start loosening some of the state controls on the individual.”

Otherwise, we would have various things made legal, and once the burden on the nanny state was shown to be sufficiently large, the nannies would have a built-in argument for proving wrong, the legalization of X.

Which, if you stop to think about it, seems to be Judge Posner’s reasoning in a nutshell. That is, put personal liberty aside for a second, this is simply costing the taxpayer too much money. So don’t fix the system, just strip one more layer of personal responsibility and liberty away from our society.

Ray G

John’s response to “critics.”

No we should not de-ioDize salt, etc. But that is a flawed comparison.

Let me fix it for you. Since you're pointing your comments at the "critics" you are thus in favor of the paternalism favored by Judge Posner.

Thus, by extension, I can accurately say that you would not be opposed to placing a ban on sea salt which typically comes without iodide. Sea salt is very popular for cooking, but plainly states on the label that it “does not contain iodide, a necessary nutrient.”

Now, using a little Posnerian projecting here, I feel safe in saying that very few people actually know why we need iodide in our diets. Therefore, we should not allow a salt product without iodide to remain on the market.

Talk about the cost of absorbing information. The need for iodized salt has been known in medical circles for quite some time, and yet the general public is still wholly ignorant of why they need iodide in their salt.

But you say “How many people actually use sea salt?”

Considering all of the “ifs” and “probably’s” that Judge P. gave us, it’s still an apt comparison, especially since the use of sea salt as a healthy alternative for cooking is on the rise.


Ray: Most sea salts contain trace minerals including iodide and the lack of iodine could have been the "transfat" issue of the twenties due to corpies of that era refining it down to "pure" NaCl. I suspect the health benefits of iodine has been known in some sense for many thousands of years considering its importance in history and being the root word for salary and cause of a number of wars etc. A military outfit like the Roman legions probably wouldn't go to such lengths as to provide salt to each soldier just for its taste enhancing effects.

"Table Salt
Table salt is the most common kind of salt found in the average kitchen. It usually comes from salt mines and once it's mined, it is refined and most minerals are removed from it until it is pure sodium chloride. Most table salt is available either plain or iodized. American salt manufacturers began iodizing salt in the 1920's, in cooperation with the government, after people in some parts of the country were found to be suffering from goiter, an enlargement of the thyroid gland caused by an easily-preventable iodine deficiency. People require less than 225 micrograms of iodine a day. Seafood as well as sea salt contains iodine naturally and the supplement is unnecessary if there are sufficient quantities of either in one's diet. Note: Natural sea salt is a healthy replacement for ordinary table salt."


The salt example seems a strong argument for "the nanny" state overseeing the integrity of our food supply and imposing regulation after the salt mfgs over-refined the natural product in the name of brand? appearance? of a commodity and making the claim of "When it rains it pours". As you mention, given the "ignorance" of the people they could have gotten away with slipping iodine out of the salt without them knowing for a long time. (Perhaps they'd later sell it back to us? As a pricey and necessary supplement?)

BTW my guess is the disclaimer on sea salt is one of those perversions of labelling that having not added iodine or perhaps not knowing how much is in the natural product a warning is in order.

There's a similar goofiness in that drug-ridden, mutant "farmed salmon" raised in pens where their feces mix with the "organic" (mostly land-based) feed pellets bear the "organic" label while the wild ocean runs of the Pacific can not carry the organic label due to our "not knowing what they feed on" in the pristine waters of the North Pacific. What would you think of a mandatory label on farm salmon warning that they have virtually none of the healthful Omega III oils associated with many health benefits and that only wild salmon are rich in? Jack

Ray G

There are links to IQ and sufficient iodine intake in a child's most formative years. Of course, the NY Times doesn't really believe IQ exists, so that angle doesn't get much play.


Posner, once again you have the better argument over Becker.

It is nice to hear someone who actually understands and respects libertarian arguments nonetheless defeat them when appropriate--

and defeat them as articulated by such a worthy proponent as Becker.


مركز تحميل


بنت الزلفي


Thank you, you always get to all new and used it
شات صوتي






Thank you, you always get to all new and used it



Very interesting site. Hope it will always be alive!


How are you. There is nothing more dreadful than imagination without taste.
I am from Zimbabwe and bad know English, tell me right I wrote the following sentence: "ˆ; Finance cost and amortisation through the accounts ˆ; accounting themes to be applied to any debt finance ˆ; accounting for the asset ˆ“ where a bond."

With best wishes :), Adiel.


I bookmarked this link. Thank you for good job!


thanks to tell me that,i think thats so usefully----
links of london

The comments to this entry are closed.

Become a Fan

May 2014

Sun Mon Tue Wed Thu Fri Sat
        1 2 3
4 5 6 7 8 9 10
11 12 13 14 15 16 17
18 19 20 21 22 23 24
25 26 27 28 29 30 31