« The Economics of Chocolate | Main | Ordem e Progresso, part II »
Should we ban trans-fats?
Gary Becker does a quick cost-benefit analysis in his head:
With a small taste benefit from the use of trans fats-- the New England Medicine Journal article I cited earlier does admit positive effect of trans fats on "palatability"-- the total cost of the ban would equal or exceed total benefits. For example, suppose 1 million persons on average eat 200 meals per year in NYC restaurants with trans fats. If they value the taste of trans fats in their foods only by 35 cents per meal, the taste cost to consumers of the ban would be $70 million per year. Then the total cost of the ban would equal the benefits from the ban.
If you click on the link, you'll see some good arguments against paternalism as well.
Posted by Tyler Cowen on December 29, 2006 at 07:24 AM in Food and Drink | Permalink
Comments
I don't often agree with GB, but I do on this one. Paternalism at its worst. As an aside, however, don't the arguments here - Becker's, and those he opposes - rely on interpersonal utility comparisons? They seem difficult to avoid once you hit the real world.
Posted by: tom s. at Dec 29, 2006 8:54:42 AM
I found it more interesting (and surprising) to learn that Posner favors trans fat bans. Is this sort of paternalism common for Posner?
Posted by: DK at Dec 29, 2006 9:40:37 AM
The cost, in terms of lost freedom and autonomy, of blocking competent consenting adults from engaging in a 35-cent transaction is far far greater than 35 cents.
Posted by: KipEsquire at Dec 29, 2006 9:55:17 AM
Eating trans fats because new drugs will offset that later? Rubbish, psychologically and medically. Lipitor does not substitute for eating healthy food! Some replacement of trans fats in products when it became known they were unhealthy? This indicates not substantial consumer knowledge but a decent ratio of (consumer knowledge * how much consumers care) / (marginal cost). Considering surveys, this indicates to me mostly low marginal cost.
Further, there is a sort of transaction cost in consumers finding food without trans fats. Finally, I greatly doubt that consumers value trans fats at as much as $.35 a meal. Hydrogenation is done not for taste but consistency, and is totally unnecessary now that (rather cheap) (starch based iirc) thickeners exist.
This borders on misinformation.
Posted by: bhauth at Dec 29, 2006 9:58:24 AM
Becker is using a $100mm benefit estimate and a $30mm estimate of direct costs to restaurants, and arguing that if the marginal "taste value" of trans-fats is more than $70mm (200mm meals x 35c/meal) the ban cannot be justified on a cost-benefit analysis. But it's virtually unimaginable to me that the taste value of trans-fats in foods averages anything like 35c a meal.
The overwhelming number of those 200mm meals are fast-food-type meals costing a couple dollars, and the portion of the meal cost to the consumer that relates to products that could be prepared with trans-fats (that is, excluding sodas, ice cream, etc.) is even less. 35c would be an enormous chunk of that residual number, particularly given that it is generally agreed that the taste difference with trans-fats is very small if it is even detectable.
I would further note (as Becker admits) that the $100mm benefit estimate is very much a lower bound. Posner gives a $3.5bb benefit estimate, and if that number is right, then the lost "taste value" per meal that would be required to make costs exceed benefits would be something like $17.50.
I think the key point here is the point Posner makes in reply: "trans fats seem exceptionally dangerous--almost in the category of poisons." As a general matter, I don't want the government rifling my restaurateur's cupboard either, but trans fats are so unusually bad as to be on a par with adding nicotine to your french fries.
Posted by: alkali at Dec 29, 2006 10:18:51 AM
As bhauth noted above, trans fats don't add to taste. The people who will be eating trans fats are the very ones in positions to know the least about their damaging effects. I expect better than this misleading cut and paste from Tyler.
Posted by: eriks at Dec 29, 2006 10:44:35 AM
A quick cost-benefit analysis in his head that (surprise!) comes out supporting what ex anti that Becker would support beforhand- what a surprise! This is a pretty typical method for him, and one that gives economics a bad name as ideological and unemperical.
Posted by: Matt at Dec 29, 2006 11:11:34 AM
Trans fats are seriously bad. Avoiding them is very, very hard and requires knowledge that is very difficult to obtain. I've attempted to avoid trans fats and even finding someone at a resteraunt who has any information about which items have them is very costly in terms of my time and the time of people at the resteraunt.
Posted by: Michael Foody at Dec 29, 2006 11:23:05 AM
I'm looking forward to the re-emergence of lard. That's what trans-fats were originally meant to replace. And it tastes better.
Posted by: Tony at Dec 29, 2006 11:25:30 AM
Count me down for paternalism. I would like to have a nanny as well.
Posted by: theCoach at Dec 29, 2006 12:01:14 PM
"I'm looking forward to the re-emergence of lard." Bread and dripping as health food! How long before Woody Allen's Sleeper comes to pass?
Posted by: tom s. at Dec 29, 2006 3:47:21 PM
Posner and Becker don't agree on this mainly because Becker doesn't seem to buy Posner's insistence that there are serious information asymetries regarding trans-fat, and the costs for learning about trans-fars so high, that people are rationally ignorant of the dangers of eating foods with transfats. Becker didn't seem to engage Posner on that point about the costs of learning about transfats.
Wouldn't a tax on trans-fats be better than a ban? Why then does NYC favor a ban?
Posted by: Jason Voorhees at Dec 29, 2006 3:47:50 PM
of course there is a search cost to finding foods without trans fats and if you value your health you willingly pay such search costs. if not...you eat at mcdonalds, you get fries with your steak instead of salad, etc.
the value of 35c does not have to be explicitly incorporated into the price but may reside in the consumer surplus. there are plenty of things that i "value" more than the cost of what i pay for and would willingly pay more.
consistency separated from taste? you're telling me that soggy french fries "taste" the same as crisp french fries?
there is no tax on trans fat because it would be highly regressive. the horror - making people that incur the cost actually pay for it. that people know so little about it is a direct indication of the relative value to them.
fat, drunk, and stupid is a wonderful way to go thru life.
Posted by: ptkelly at Dec 29, 2006 6:28:29 PM
I believe the ban on trans fat is far more defensible than the ban on smoking. It's easy for me as a consumer (or even a worker) to tell whether a place allows smoking and make my decision accordingly. It's very hard to know whether the restaurant uses trans fat or not.
And a tax might be ideal in most cases, but here the administrative burden seems pretty high.
Posted by: Keith at Dec 29, 2006 6:35:25 PM
The underlying argument to the ban, inasmuch as any of it can be reasonable, is of course that it is difficult for the consumer to learn about trans-fat, and then equally diffficult to detect when it is present in their restaurant purchased food.
Having said all of that, it is the height of arrogance to tell them what they can, and cannot eat. To think we have people fighting for the legalization of drugs, prostitution, euthanasia, all in the name of personal freedoms. But God help you if you use lard in my tortillas (and yes it makes a difference - you should taste the home made tortillas we buy).
Has anyone stopped to consider the black market forming for Crisco? Organized crime, street gangs, all engaging in rampant crime sprees in search of the inflated profits that illegal trans-fat will surely bring on the black market.
"Psst, buddy, got some Crisco in the trunk, uncut, best price. . . "
Posted by: Ray G at Dec 29, 2006 7:35:31 PM
Re-read my comment, then save it for the war on drugs.
Posted by: bhauth at Dec 29, 2006 9:59:49 PM
There are surely some positive “knowledge” externalities for the rest of the country (world?) we should factor in. Local restaurants experiment a lot more than usual, and sooner rather than later, better tasting and healthier substitutes are bound to be discovered. In the concrete example of NY, a number of fast food chains have now been experimenting with soy-based oils for a year or so. Surely this will be good for consumers everywhere else: if you live in VA you will soon have more choice than ever, without being FORCED to part with trans-fats.
Posted by: silviu at Dec 29, 2006 10:16:55 PM
The story of decaf coffee just came to mind.
Certain cereal makers and health-nuts started a campaign against coffee, claiming it was essentially poison, about 80 or so years ago.
"Big-coffee" countered in a number of ways, one of which was to push decaf as a viable alternative.
Transpose this situation to those times, or vice versa, and we would essentially have NYCity banning coffee.
Posted by: Ray G at Dec 29, 2006 11:46:29 PM
Ptkelly, you get soggy fries without trans fats if you are being cheap and or incompetent about cooking them. Trans fats lowers the skill and resources needed to make good fries.
Posted by: TW at Dec 29, 2006 11:47:12 PM
bhauth;
Relax, no one was talking directly to you, until now, and it was a joke anyway.
Re-read this comment and take some drugs, . .. oh wait, I mean re-read my comment and, . . .hold on. . .
Posted by: Ray G at Dec 29, 2006 11:50:40 PM
Posner: “They [the consumers] have, in short, no idea of the benefit of avoiding trans fats in restaurants.”
Therefore, the government is correct in banning a substance that is, and I quote “(very probably) harmful to health.”
This follows the very subjective reasoning that the average consumer is capable of understanding price differences in one commodity, but says Posner, they probably wouldn’t be able to understand the whole trans fat issue.
In a nutshell, we see the inherent problems in trying to plan a society, for whatever reasons. One man or group of people making broad and subjective assumptions about the rest of the population.
Posted by: Ray G at Dec 30, 2006 12:50:05 AM
Ah, OK Ray. But I have seen people seriously argue the equivalent of that.
Posted by: bhauth at Dec 30, 2006 1:14:00 AM
"Ah, OK Ray. But I have seen people seriously argue the equivalent of that."
I thought the Crisco in the trunk line would have given it away. And I don't think Crisco is even considered a transfat anyway, but I couldn't think of an actual name brand source of trans fat.
The bottom line is still the govt making a decision for me (in principle since I don't live in NY) that really isn't theirs to make.
My main argument against most libertarian proposals is precisely the existing paternalism in our present soceity. We would essentially be setting the legalized thing, call it X, in a place of certain failure.
Once the externalities of others' reckless behavior was realized, the "nannies" would have a great argument for saying why X was illegal all along. Whereas if we fix the system first, and then begin to loosen the govt controls on our individual liberty, we will be walking down the hill to get all of the cows, instead of running down to get just one (to cite the old joke about the young bull and the old bull).
Posner still claims he has a libertarian, Millsian position in this because of the externalities that trans fat related heart disease transfers to the taxpayer. But he is coming at it from the wrong direction.
Not legalizing X for the aforementioned reasons is simply being prudent, i.e. in the long run, X is still made legal. But Posner's argument is to make illegal practices that are currently lawful, or were. This moves in the opposite direction of a freer, Millsian model, by basically giving in to the existing paternalism of our current system, and thus allowing that paternalism to shape future decisions about liberty.
Posted by: Ray G at Dec 30, 2006 1:53:32 AM
The word "paternalism" should never leave the lips of an economist. Its use is a dangerous sign that the economist has allowed analysis to take a back seat to ethos via labels.
Posted by: Tree at Dec 30, 2006 1:27:48 PM
A few bones of contention:
I believe that most people are ignorant to the health hazards of trans-fats. Couple this with the fact that these cooking oils are more cost effective for the food producers and the likelihood of the market stepping in to fill the void for trans-fat free restaurant food is pretty much nil. Furthermore, even a trans-fat wary individual has a near impossible time steering clear of the substance in a restaurant setting as it's not listed on menus or ingredient lists.
These health hazards effect a cumulative cost not only on each individual consumer, but eventually imparts an external cost on society as a whole (early mortality/morbidity, diminished productive capacity of the individual, increased health care costs).
Is it paternalism to discourage drunk driving? What's the cost measure of pleasure that the would-be intoxicated drivers are missing out on? Does that constitute anti-libertarianism?
$0.35 of pleasure per serving? I find that difficult to believe. However, if the cumulative depravation cost is too great for society to bear, how about removing the trans-fats from the food as served and those that are willing to bear the cost trade-off can purchase spray-on packets of trans-fats to add to their fries etc, complete with a warning form the surgeon general?
Posted by: kebMo at Dec 30, 2006 1:40:47 PM
