« Should We Worry about the Rising Inequality in Income and Wealth?--Posner | Main | Comment on the New York Ban on Trans Fats--BECKER »



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


Let me start with an obvious libertarian objection. Even if your math is correct, if this ban happens, it will greatly encourage more such laws, where government limits the power of indviduals to decide how to live their lives. And, I should add, it's doubtful these future laws will be as well-calibrated economically as the trans fat one.

So how much is this cost? Hard to put a number on freedom, but let's conservatively estimate it's worth several billion for NYC per year.


I don't buy the slippery slope argument. That was advanced twice already, once with economic regulation in the 30's and once in the 70's during the rights revolution. "If we regulate safety equipment in factories, soon we'll regulate everything." That hasn't happened there and there is no reason why it should happen here. Don't forget, even NYC has something of an elected government, and if people don't want to exercise at specific times, they can vote fat slobs into office. We have ways and means to prevent the slippery slope.


So you think we should reject the slippery slope argument because two times out of thousands of potential cases it hasn't worked? (Actually, in both cases you mention we've had massive overreach that should make one question if either was successful, or at least ask if greater regulation was the proper way to handle these situations.)

Read Eugene Volokh's article on slippery slopes. It may open your eyes.


That was a fun analysis! I am confused, though.

You avoid the insurance math because you say that elimination of heart disease "would result in an increase in illness and death caused by other diseases, such as cancer." That makes sense.

But given that fact, isn't it inappropriate to count all 500 people who won't have a deadly heart attack as saved from death for that year? 88% of them are 65+ years old...

Or were you including that when you said that the elderly have below average value of life?

In any case, your more conservative 100 lives saved per year probably sweeps probably takes care of all this.


I think that a middle option might be the best. The FDA's labeling requirement for trans fats has significantly reduced the number of store bought foods containing trans fats. Once labeling was required, people have begun to take notice of the presence of trans fats and discuss what they are.

You state that "no one wants his restaurant experience poisoned by having to read a menu that lists beside each item the number of grams of trans fats it contains." The labeling need not have to be so intrusive. Many restaurants already use an asterisk to indicate that something is spicy. Some similar universal symbol could be used for the presence of trans fats. I do not think the precise number of grams needs to be stated on the menu (although it should be available upon request). This will not "poison" the restaurant experience for most individuals. It will get people thinking about the issue and will cause some change. Once people know that the food contains trans fat, they can make their own decision about whether or not they want to order that dish.

Peter Pearson

Much of Mr. Posner's argument is the observation that it's economically inefficient for each individual to think about trans fat, and that costs are lowered by having the government do that thinking for us. While I'm personally inclined toward the opposing, libertarian position, I would like to propose a compromise: Take all your nanny-state rules and bundle them into a single Nanny State Seal of Approval, which could be displayed by establishments that choose to follow those rules, and not by establishments that choose otherwise.

Laws already on the books (fraud) would be used to punish cheaters who display the Seal without following all the rules.

A particularly appealing benefit is that the fraction of businesses displaying the Seal would be a real-time readout of the economy's opinion of the usefulness of the Seal and the rules behind it. And concern for that very visible indicator would give the rule makers and incentive to keep the rules reasonable: if you walked down the street and no restaurants displayed the seal, you would know beyond a doubt that no significant fraction of the customer set considers the rules worthwhile. The bureaucracy behind the rules would see that it had lost its leverage, and would set about fixing its rule set.

This compromise captures most of the savings of having the government do our thinking for us, without our sacrificing additional liberties.


Post written for Becker Inequality....... but! as complained about before it appears the thread was closed (for some reason??) while I was typing a response. Is it not possible to post a warning? Assuming pixels costs make it necessary to close threads?

Nelson: Your ideal is interesting and to be sure, by comparison I'm taking "baby steps" from our past policies of "we stole your land by force, but otherwise "fair and square" and screwing Mexico and S-A vigorously and often over the intervening decades, but even bring the focus of the discussion of the "Mexican immigrant problem" below the border and asking that our "diplomats" seek policies beneficial to both is enough to have me labelled "soft on something or other" or "soft in the head" by the fence em off and forget em set.

But! stodgy as it may seem I've reservations as to having no nations and no borders but instead favor a "nations" or groups who can run somewhat differing experiments as we do in the US with different states. For example, I live in Alaska and it takes newcomers quite a while to learn why we might do things differently here, or better example, when our rich salmon fisheries were managed (before statehood) by those in WA DC the resource was wiped out. Today, under Alaskan management it's been restored to the runs of the early 1900's Currently there's a raging controversy over a proposed mining of a huge deposit of copper and gold, but the mining tech used for that type of mine may lay waste to Lake Illiamna (perhaps the largest and best sport fishing Lake in the world) and damage or ruin the huge Bristol Bay Salmon run. I doubt this would turn out well if it were up to a few Congressmen taking 3 day junkets; instead much of the outcome will be decided by those living here, native Alaskans in the region etc.

But more of your post is about the free migration of labor, and I'd agree that where a labor vacuum exists next to a country in labor surplus the vacuum is likely to be filled..... by one means or another.

But! there are concerns here as well. Say one country has sacrificed in some way, say to preserve their rivers, farmlands or the environment in general, while over a century a neighboring nation has laid waste to their natural resources, so when the inevitable famine or disaster strikes them, do they just pack up and move to the "green" country? And try to do the same?

Coming back to the present, fortunately, despite tremendous adversity 99.5% of Mexico's 100 million people chose to stay home each year; they like their country, family and friends. But half of them are under 30 and probably are looking at 30% unemployment in that age group. So the least disruptive policy would seem one that purposely exported some of our labor shortages to Mexico (if need be at the expense of rapidly growing China or India) in hopes of bringing their wages and levels of employment closer to ours.

If we "got it right" 50 or 100 years from now Mexico might play more of a "Canada role" in North America and freedom of choice would seem much easier. Today the problem for both "globalization" and immigration in many parts of the world is simply that the pace is too high.

Haris, also in "free trade" et al it's the pace of change that is far too high and with too many Americans being naively provincial I think we fail to "GET" just how large the labor overhang (unemployed or starving with a "job") is (something like 1.5 billion) and no matter how strong the market of our 5% of the world's population is we can NOT employ (offshore et al) all of them without taking our wages down so far that the living std of the US will be unrecognizable in the very near future.

The economics of "relative advantage" used to speak to the natural advantages one trading nation held over another at similar pay scales not the employment of .50 labor having the "advantage" over the nation of $20 labor.

There are but two choices here (other than ruin) the US has to aggressively find SOMEthing or MANY things to export or adopt policies to dramatically slow our imports. The "free trade" theory is fine but it only works in "the long run" and as Keynes famously said, "In the long run we're all dead". Jack


Posner is, as usual, dead on.

There's an added dimension that might persuade some of the objectors that this is not necessarily as bad a thing as they might imagine (slippery slope arguments notwithstanding, I'm arguing solely on the merits here).

Trans-fats were invented as a low-cost ersatz to the butter and lard that were primarily used. In the 60s adoption spread quite a bit as ADM and others pushed it as a healthier (we now know better) alternative to margarine but the primary reasons for adoption were shelf-life and cost*. As such, banning them is more akin to banning fake or watered-down milk - it's just that it took a lot longer for us to realize the con in this case.

Requiring restaurants to use the genuine article instead of the knock-off is just categorically different from mandating the Ornish diet or banning alcohol. It's more akin to the governments role as a neutral mediator in the market than any of the more radical parade-of-horribles being brought out against the move.


*Ironically, the savings are not actually all that substantial except in the very highest volumes. The 1% figure is off the mark by at least two orders of magnitude.


If what you said is true ,why not ban all cigarret or other things which perhapes do harm to people?The Chicago ecomomics not only has a very excellent reason for ecomomics,but also can guanrantee our freedom.So don't simply critisize the Chicago economics.We must cauciously pay attention to any regulation of government and try our best to safe our freedom.In this case,I think the best way is to require restaurant give consumers notice about tran fats if some food has and it is absolutely not necessary to ban it.


While I agree that banning transfats is a wise, helpful thing to do, I can't resist criticizing the way in which the author reaches the same conclusion.

The writer supports transfat bans because the consumer has insufficient information to reach an informed decision on his or her own. But, imbalances in information are always present in the world. It is only in the classroom that one can deal with markets that have inlimited resources, unlimited consumers, all fully informed, etc., in theory.


Judge Posner:
Respectfully, the cost-benefit analysis misses the point.
The point is that an ever-encroaching state which incrementally insinuates itself into the lives of its citizens will not stop at a ban on smoking in bars (not coincidentally, also put in place by the same mayor), or a trans fat ban in restaurants. Instead, a nanny state which infantilizes its citizens in the guise of protecting them from heart disease will--ironically--make them less responsible for their own health maintenance. This, in turn, also makes it more likely the government will continue to arrogate for itself ever increasing powers, including in areas unrelated to health. And, at that point how will the populace be able to object?


I like Peter Pearson's proposal. Have an official Healthy Seal of Approval. That way people can tell quickly and easily if a particular meal passes the standard government healthiness regulations, while still maintaining the freedom to create and consume unhealthy foods.


If the cost-benefit analysis simply included all the easily measured costs (hospital care, restaurant expenses, etc) then it could be quickly countered by the argument of libertarians that given these risks, people should still be allowed to choose to eat trans fat. After all, we allow people to eat red meat, sky dive, and all other sorts of stuff that are potentially harmful (and costly). Presumably people do these things because they make life more enjoyable.

However, the key aspect of Posner's analysis is that it also incorporates the value people put on their own lives. This value is calculated from what kind of risks people are willing to take for such activities as eating red meat and sky diving. Thus, if the ban results in a significant net societal gain after taking the value of life into account, this suggests that consumers are not correctly evaluating the risks associated with trans fat as well as they do the risks associated with other activities.

As for the slippery slope argument, I answer is this way: I agree that we must remain ever vigilant to protect our rights and keep the state from unnecessarily interfering in out lives. However, I think this case is in the same category as prohibiting poisons in food (which most would agree are good laws), rather than prohibiting smoking. The reason is that, while many people enjoy smoking even after being educated on its dangers, I know of no one who is particularly attached to trans fat. The only attraction is the marginally lower price. But we have seen that people, if educated, would not accept this risk in order to get the tiny price savings.

(Thanks Posner and Becker for the fantastic blog.)


Rather than ban trans-fats outright, why not tax them in the Pigovian style?

Ben G

Posner's arguments rests not on the negative externalities of consuming trans fats (though he does consider the possibility) but rather on information assymetry and consumers' inability to make rational decisions. I'm skeptical of Posner's admittedly tentative cost-benefit analysis and also of the idea that, assuming Posner's analysis is accurate, a ban is the most effective policy.

I have not read the literature on trans fats, but it seems it would be very difficult to pick out a number of deaths that are "caused" by trans fats. Are the health costs of consuming trans fats noticed in the short term or long term. That is, do the 100 and 500 deaths per year figures include deaths 20 years from now that were partly caused by trans fats consumed this year? If they do, then they depend on the assumption that no new medical technology will be able to prevent those deaths. If they don't, then the real purpose of the policy should be to prevent people from eating trans fats only if they already have a reasonable risk of heart attack--a significant but not overwhelming percentage of people.

The $7 million number is based on an average, unbiased sample. But people who consume a large amount of trans fats presumably demonstrate a propensity towards risk, so that number should probably be decreased somewhat.

But even if the cost-benefit analysis works out as Posner suggests and information assymetry is seen to be the cause, I think a ban is a poor choice to improve the situation. I would prefer an excise tax on trans fats, which would add an explicit cost to the sale of products that use trans fats to account for the hidden cost of their consumption. Better yet, at the same cost as enforcing the ban or excise tax, an aggressive advertising could provide the information to consumers that would get them to place pressure on restaurants. Think of how the quickly the restaurant industry changed in the wake of the Atkins/South Beach fad. No government forced fast food companies to offer wraps and focus advertising campaigns based on how few carbs their products contained; it resulted from a (somewhat misguided) demand for low-carb foods.

The Emperor

I wonder if the restaurant industry was secretly happy with the outcome. It's very possible that what they really feared was a broad health labelling requirement for all restaurant food. Compared to such a requirement, a ban on trans fat is not that bad as it keeps hidden the relative unhealthiness of restaurant food. Actually, it makes them look good -- they can now proclaim they are trans fat free and thus healthy!


I was considering making a case of lower health care costs offsetting any real or imaginary costs of banning oils that have been chemically altered to enhance shelf life and profits. After all the costs of heart disease are not limited to the early deaths but added costs before heart surgery or death, and the consequent loss of individual enjoyment of health, and loss in productivity.

But then I remembered how medicine "works" in the US and that any decline in the number of heart surgeries would most likely result in a price increase "to cover the lost revenue" and that the cost of "mistake" insurance would be spread over fewer invoices, and some $100 million insurance CEO would justify a bump in benefits for having had to manage the "situation".

I get another chuckle out of the idea that perfect knowledge would combine with "market decisions" for the benefit of all within anyone's current life time. Singlely? maybe. But in vary groups of diners? After agreeing a spot for proximity, ambience/tolerabitity, price and menu, a voice pipes up from the back seat? or cell phone that his-her "handbook" indicates the place is still frying spuds in "bad oil" and putting it in their breads and desserts.

For overly "religious" Libertarians, might you settle for an opt in? Say being allowed to ask for your french fries in corporate designed oil? And leaving it to the market to decide if they want to cater to you?

As for Nelson's suggestion of a "official Healthy Seal of Approval." are we likely to agree on what constitutes the "approval?" Case in point for salmon shoppers: Wild, healthy pacific salmon laden with healthful Omega III oils are not labelled "organic" because "we don't know what they eat" in the pristine waters of the northwest, while pen raised fake salmon ARE to be "organic" as we KNOW what sort of land based "organic" garbage is fed to them.

Ben sez: (along with others) "I would prefer an excise tax on trans fats, which would add an explicit cost to the sale of products that use trans fats to account for the hidden cost of their consumption." Again..... the study of econ and design of policy is not entirely the cold calculations of various econ graphs and models and the "young" in this case those who've not had serious health problems, yet, rarely, if ever know or act on the costs of their actions. Is there anyone at all who does not know the cost of driving fast and tailgating yet it is far more common than not. Is there a 20 year old who "thinks" he's mortal? And is there a "reason" why a military draft is aimed at 18-21 year old kids instead of say 22-28 that would provide the more mature and experienced military as won WWII?

Best for the Holidays, and remember we are first a democratic community humanity that selects the best of capitalist principles for OUR use not the opposite. Jack

Christopher Clock

The NYC government has had dramatic success in recent years with top-down ordinances such as the smoking ban. Although there was much initial bellyaching, the silent majority overwhelmingly approved after some time had passed and people had become accustomed to it.

Smoking has significant negative externalities, but nobody wants to ask another to not smoke. Every restaurant found it in their best interest (if only slightly, and by dint of custom) to allow smoking. Here, political action solved the collective action problem.

Similarly, restaurants find it only slightly within their best interests to use trans fats. People don't know which do and which do not; they would vote against if they were sufficiently well informed, but it's not enough of a problem for everybody to become informed about. The government steps in and solves the collective action problem yet again.

Many of you see any government reach into our lives as intrusion by some "other." But you elected the government, and you can kick it out. If you had the option to not eat trans fats, you would. Government is here to solve problems by way of enforcing optimal solutions.

The true slippery slope goes the other way, when folks find it unpleasant to ask others to put out their cigarettes.

David Welker

I don't see why Posner buys into the idea that human life can be translated into money. This seems like a foolish idea. I don't think the average person would sell themselves into slavery for $7 million, much less sell someone else a license to kill them.

The idea that if a person is willing to do a more risky job, say one that pays $n dollars more with a risk increase of say, 1% that we can then go from there to a "value" of human life is ridiculous. First, just as information costs prevent people from assessing the increased risk from trans fats, it obviously prevents them from assessing the increased risk from, say risky jobs.

I know that working in a coal mine is more dangerous than some other jobs. Do I know how much more dangerous? No. I don't have the statistics. Do most coal workers have the statistics? Clearly not, at least if they are anything like most other Americans. So, how does one translate the higher pay one gets from say, working in a coal mine to a value of life when mine workers do not even know the risks they are taking?

Second, even if someone was willing to take a 1% risk of death for $n, that doesn't mean they are willing to take a 99% for 99 * $n. It is an absolute certainty that this function is not linear. A 1% risk is quite remote. Most people, when confronted with this risk, are likely to irrationally assume they will be safe. Or alternatively, overestime the risk and freak out emotionally, avoiding the work in question all together. But if you give them something like a 99% chance of death, you are making death a near certainty. Those who underestimate a 1% risk are much more likely to realistically assess a 99% risk.

How many people do you think would accept a 99% chance of death for $7 million? Not too many. Only those who are desperately unhappy with their lives due to financial reasons would accept that deal. Or who value their lives very little (perhaps because it is very painful or will not last very long anyway), and want to leave something to family members. So, if the vast majority of people wouldn't accept that deal, they obviously woudln't accept a 100% chance of death for $7 million either.

A more realistic view is this. Life cannot be exchanged for money. They are on different dimensions. Money cannot buy everything. It really is not sensible to pretend otherwise.

It is NICE to think that all the difficult questions can be reduced to cost benefit analysis and thereby rendered more objective. Unfortunately, this is simply not the case. There is nothing objective about a $7 million figure for the value of a life. Thus, there is nothing objective about a cost benefit analysis that depends on that value.

I agree with Posner's conclusion. That banning trans fats makes sense because people do not rationally assess the risks they are taking when they go out and eat trans fats in a restaurant. But my conclusions stops and rests on the irrationality (or rational ignorance, if you prefer) of consumers; not on some artificial cost benefit analysis.

Julie Fitzgerald

There are two things wrong with the cost-benefit analysis conducted here. First, medical interventions are rarely coupled with cost benefit analysis for a reason the author pointed out. Benefits start spiraling out of control and it is difficult to compare costs that are so different. For this reason, it is more beneficial to conduct a cost effectiveness analysis. Secondly, when calculating benefits, it is important to recognize that the heart attacks prevented aren't all fatal ones. This helps to decrease the high level of benefits that we are seeing in this analysis. About 40% of heart attacks are fatal. The rest certainly have costs associated with them, but much less cost than mortality. Using the high rate of a statistical life is also an issue. It would be unwise to assume that the life expectancy of someone who suffers from heart problems is the same as someone who doesn't. In fact, research has suggested that those who suffer from a fatal heart attack are losing, on average, about 11 years of life. This reflects their health conditions and the long term probability of good health.

Also, the effect of a trans fat ban is dependant on the percent of trans fat intake related to total dietary consumption. Generally speaking, a 1.5 percent decrease in trans fat decreases a persons risk for heart attacks by 1 percent. This decrease does not occur immediately. In fact it takes about 3 years for the effects of a decrease to be beneficial to a persons health.

Finally, while many have suggested that this ban would decrease consumer surplus and somehow result in dead weight loss to both consumers and producers, I think it would be fair to argue that these losses are captured and overshadowed by the associated health benefits.

This ban, and others like it simply reflect an increase in knowledge about the impacts that consumption has on our bodies. It is well within the rights of a public health organization or the FDA to ban things that have adverse affects on our health. In fact, I am quite certain that that is specifically what they were created to do. When a drug has been shown to have adverse effects on our health, you don't hear people talking about personal rights and government intrusion when it is banned. This should be no different.


As for Nelson's suggestion of a "official Healthy Seal of Approval." are we likely to agree on what constitutes the "approval?"

The seal just means they follow the "no trans fat" and similar regulations. It's an easy way to see if the food we want is healthy for us, according to the government, and not be fined or have our favorite restaurants shut down if we want to make or eat something unhealthy. If you had read what Peter Pearson had posted, you'd know that it wasn't really my idea, I just thought it was a really good one.


Sure, this is a government regulation. But I think a more appropriate analogy would be sanitary standards. I presume even the most libertarian prefers restaurants that don't have a huge rodent and roach problem in the kitchen, so no one objects to periodic health inspections. Pest control services for restaurants aren't cheap, either.

Half Sigma

Sure, this is a government regulation. But I think a more appropriate analogy would be sanitary standards. I presume even the most libertarian prefers restaurants that don't have a huge rodent and roach problem in the kitchen, so no one objects to periodic health inspections.

Good example of a case where there's assymetry of information. The restaurant owner knows how dirty or clean is kitchen is, but the diners don't.

Government intervention solves the informational assymetry propblem.


What will the restaurants use in the place of trans fats? If the answer to that question is saturated fat or a non-trans partially hydrogenated fat, then the health benefit might not be all that great. Saturated fat is nearly if not equally as bad for your cardiovascular system as trans fat.

Labeling seems like a good idea regardless of the ban. I want to know what is in my food, not what isn't.

Stephan Johnson

The problem with Posner's use of information costs is the heart of the matter (as some posters recognize), but I think a crucial bit of analysis is missing. Informational assymetry is something that, in and of itself, has its own reward and cost schedules. That is, the whole point of education is to incure a cost to put oneself in a better position vis-a-vis information. Hence, assymetry in information, for all its costs, is an essential motivating factor (if not a conceptually necessary ingredient) to education itself. By trying to iron these out, at a substantial cost to those not necessarily in the market for that information, we lose the value of assymetry (in motivating and rewarding education) in the first place. This is absolutely parallel to redistribution schemes in other areas: by attempting to remove assymetries, we create worse problems than those we were trying to solve. And as a factual addendum, how costly is it really, given the internet, to know things like 'trans fats are bad'? Once again, the internet proves to be a much more efficient means of distributing assymetries and rewarding those who are motivated by them to learn more.

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