Nutrition Labels

Requiring food nutrition labels doesn’t get people to eat healthier:

A study of New York City’s pioneering law on posting calories in restaurant chains … tracked customers at four fast-food chains — McDonald’s, Wendy’s, Burger King and Kentucky Fried Chicken — in poor neighborhoods of New York City where there are high rates of obesity.  It found that about half the customers noticed the calorie counts, which were prominently posted on menu boards. About 28 percent of those who noticed them said the information had influenced their ordering, and 9 out of 10 of those said they had made healthier choices as a result.  But when the researchers checked receipts afterward, they found that people had, in fact, ordered slightly more calories than the typical customer had before the labeling law went into effect, in July 2008. (more)

A new study involving the Taco Time fast-food chain in Washington State.  Researchers found that adding calorie counts to restaurant menus had no impact on diners’ choices.  Similar studies in New York City have recently reported conflicting results; some surveys showed that menu labeling led to fewer calories purchased, but others found no difference in meal selection. Researchers are not discouraged by the results, however, noting that providing nutritional and calorie information on menus may still benefit consumers indirectly.  As more local authorities mandate such changes, food vendors are pre-emptively modifying their menus to both lighten existing options and add healthier foods.  (Time 1/31/11, p17)

So the reason to require food nutrition labels is to scare producers into offering healthier food, with the implicit threat that if producers don’t fall into line stricter regulations will soon follow? Really? Seems to me this is more driven by a public opinion unwilling to update on the evidence. Ordinary people think labels should help, so support label laws, and the rest of the policy and political process just falls into line.

Added 31Jan:  A new paper says calorie posting at Starbucks reduced average calories per transaction by 6%.

GD Star Rating
Tagged as: ,
Trackback URL:
  • James

    I know anecdotes are weak, but I certainly noticed the labeling on McDonalds food, on the actual package, not on the menu. Now I don’t eat fast food often (twice in the past 4 months) and always out of necessity, but it certainly influenced my food choices. No more soda and fries for me. It didn’t help the first time, but it definitely affected the second and all future times I eat there.

    Of course I still maintain if you really want American to be healthy, just make gym class actually fail-able for fat kids.

    • Considering the activities involved in most school’s gym classes, I don’t think that would make much difference. Besides, no amount of exercise is going to make up for eating the McDonald’s diet when the kids go home.

  • I don’t expect much more from “normal people” as far as updating their common sense beliefs in light of new evidence, but it’s troubling that “researchers are not discouraged by the results” considering that their whole focus is supposed to be the evidence.

    • Since when has regulation ever been evidence based?

      Since the law went into effect, July 2008, there has been a pretty severe recession. Recessions cause stress, particularly among poor people, stress causes people to eat more. Poor people eat where food is cheap. The observation of no change might be an improvement over what would have been expected.

      The study only looked at time periods two weeks before and two weeks after the regulations went into effect. That may not have captured any change.

      • Note that I referred to researchers, not regulators.

  • dave

    We don’t expect dregs to care, right?

    Only people who already care about health would process that info and change their purchase, and those people already don’t go to McDs with any regularity.

  • Uh. You’re talking about a test conducted entirely within McDonald’s, where the product is made of fat, sugar and salt with a slightly proteinaceous substrate. The stuff is made of ill health. I find it implausible that you can draw a conclusion like “labels don’t matter” from a group preselected from those who pay no attention to the nutritional value of what they’re eating. A supermarket or somewhere that actually sells food would be way more convincing.

    • Jordan


      Nutritional labels are valuable not because of their influence on eating patterns, but because people who care ought to be able to know.

  • Robert Koslover

    Please, for those of us with multiple and sometimes quite-serious food allergies, the availability of detailed ingredients labels (the more detailed, the better) is extremely important. And although the “nutrition” breakdown is considerably less important, it is still surprisingly useful in providing critical clues about the contents of some of the less-than fully-disclosed ingredients (like “natural flavors”) that appear extremely commonly in ingredients lists. Yes, this has almost nothing to do with nutrition per se, but it still matters, and not just in regard to allergies. There are also many people who are diabetic and must limit their sugar and carbohydrate intake, or who are seriously gluten-intolerant, or who absolutely must avoid aspartame and similar compounds, or who need to keep their sodium intake very low, etc. The ingredients and nutrition labels (when accurate) are essential tools for them. In my own case, if I don’t know (or have pretty high confidence) about all the ingredients in any particular food, I simply don’t eat it. That applies to most items that appear in buffets, for example. Now, as an “axiomatic libertarian,” I generally don’t favor the heavy-handed imposition of laws requiring the posting of ingredients or nutrition lists. But I do strongly encourage all who work in food-related businesses to provide such lists voluntarily as a valuable service to their customers. Thank you!

  • Curt Adams

    But when the researchers checked receipts afterward, they found that people had, in fact, ordered slightly more calories than the typical customer had before the labeling law went into effect

    That is possibly the *worst* control group I’ve ever seen. They’re comparing different groups, one highly selected (labeling readers), at different times.

  • James Babcock

    When a food is advertised as low-calorie, I don’t think “healthy”, I think “small serving size”. To decide whether something is healthy, I look at the macronutrient ratios instead.

  • MPS

    Isn’t it a sufficiently good reason that some people want food nutrition labels? Isn’t the lesson that from economics that food nutrition labels should create a more efficient market?

    I’m trying to understand why anyone resists this idea, other than because they perceive the requirement as somehow diminishing the status of business people, while elevating the status of regulators, and because they think business people deserve more valorization, and regulators less, they instinctively respond against.

    (Of course your post-hoc rationalization will be that producing food nutrition labels is a burden to food providers. But I think if you face up to this objectively, you see it’s not a big deal. I figure it’d take one person less than 10 hrs with a recipe list and the internet to figure out the necessary information. Compare that to your market-theoretical advantages and the fact that many consumers want it.)

    BTW it could be that people determine to eat what they want to when they eat out, but with nutritional information in mind adjust what they eat before or after at home. In fact I know this happens, because I’ve done it before.

  • Sister Y

    Do people who eat fast food even know what calories mean? Or how many they burn in a day? Or how many calories adds up to a pound of fat?

  • Doug

    People at McDonalds are innumerate?! What a shock!

    What are you going to tell me next, that pro-exercise messages in operas have no impact on public health?

  • JGWeissman

    If people ate at McDonald’s less often as a result of seeing the labels, would the study have noticed?

  • IMO we do not know what foods are more healthy than others for people who get sufficient vitamins, minerals and proteins which includes 95%+ of Americans and Western Europeans. Consider the more dietary fat less dietary fat debate and the lack of convincing data.

  • Zvi Mowshowitz

    I know that I have benefited directly from seeing the labels on foods and have used them to alter my food choices, and I know others that have done the same. It is absurd to think that they ‘don’t change behavior.’

    There are any number of ways for this effect to be large and yet not to result in the average customer at a given restaurant consuming less calories. Perhaps they consume them in the same quantities but get what they would most enjoy. Perhaps those who care now go less often, and are counted less in the average. Perhaps once people know what they’re buying the effective price goes down so they consume more without the cost of uncertainty. Perhaps many of the customers at McDonald’s weren’t looking to consume less calories, and some may actively have wanted to consume more for less money, in which case good for them. Perhaps as noted above overall conditions have evolved.

    I don’t expect I would on average consume dramatically less calories when I go to a given restaurant when I had the numbers in front of me, but I do know I would happily accept substantially higher prices to get that information.

  • Jim Stone

    One purpose of food labels is to try to cause those who don’t pay attention to how they eat to start caring.

    Another purpose is to allow those of us who do care what we’re eating to know what’s in our food.

    I’d like to see labeling requirements go even further than they do now. Not for the first, paternalistic, reason, but for the second, full-disclosure, reason.

  • Doug Winter

    You are drawing a lot of conclusions from a single, small study…

  • I prefer a rational consumer explanation. People already know fast food is loaded with calories. Labelling does not tell them anything they do not already know.

  • Michael Kirkland

    If you believe markets make good decisions, certainly you oppose hiding information from markets, no?

    It seems to me that the fear of “sticker shock” is the force that would drive the restaurants to alter their menu, not some nebulous unmentioned future regulation.

  • I had a feeling of deja vu, since I got into an argument with Hopefully Anonymous (linked from here) on a similar study months ago. Sure enough, it’s the same one.

    James, Zvi, your own self-reports aren’t terribly solid evidence. The study showed that many people believed they ate healthier as a result, when their receipts showed otherwise.

    David Gerard, did you not read the post? The were other restaurants (Wendys, Burger King, KFC,Taco Time). I don’t know if any of them are that much healthier, but it’s not “entirely within McDonalds”.

    Robert, I wasn’t aware some people had aspartame allergies. Not that it’s particularly surprising, just wouldn’t have thought of it myself. Something to keep in mind.

    If anyone wants links to actual studies rather than reporting, I provided some here. As I recall, Curt, they gave results for people who are not label readers as well.

    I’ve got a review of Tom Naughton’s “Fat Head” (made in response to “Super Size Me”, though he fortunately remade much of the film to showcase Gary Taubes’ and likeminded folks theories) in the works, but haven’t finished it. You can watch Naughton’s movie for free on Hulu.

    • Zvi Mowshowitz

      I did NOT say that I reduced my caloric consumption during a given visit or average visit. I said I benefited directly from the knowledge, and that’s a very different statement. In several cases I have shifted my consumption and gained utility by doing so. I think of calories as a budget that I have to spend.

      Think of calories as a currency. In one case, I order from a menu without prices and then pay the bill whatever it is, estimating what things probably cost. In the second case, I know what everything costs and choose what to consume accordingly. I will thus make better choices about what is worth the cost, but I may end up spending more in that case than in the first case because I can spend them more efficiently and without worry that I’ll overspend. When I have no idea what the prices (calories) are, I often get risk averse.

      There’s also an effect across venues. I liked the idea of going to Cheesecake Factory until I saw a calorie-labeled menu and realized what they were up to, so I now almost never go, but on an average visit my consumption probably doesn’t change much because otherwise why would I go there. On the other hand, I’ve learned I should be going to steakhouses more often than I was since I went to one that labeled, since I thought they were worse for me than they are, but if anything I consume *more* calories when I’m there because I feel like it’s all right. Also brings to mind that people drive faster when they’re wearing seat belts, but again even if they’re not safer they’re still better off.

      I also don’t think we can assume that the noticing and not-noticing groups would otherwise have consumed similar caloric amounts. I watch calorie counts because if I don’t watch myself I eat too much; if I was naturally skinny I wouldn’t care. I also tend to notice such counts more when I’m planning on eating things that I believe will be higher in calories, since I have more reason to bother. Of course, it could also run the other way for several reasons, such as those noticing trying to eat healthy and those not noticing being indifferent.

      I have chosen restaurants over others because I wanted to see those counts, and it weighs substantially into my decision. I would pay a substantial sum right now to know the counts on the menus at several local places that I go to or order from regularly! In fact, I’m going to ask: Is there any way for one of us to get that information, by taking the food in question and measuring its caloric content directly? For sufficiently repeated interactions this strikes me as worthwhile if prices are reasonable.

  • Pingback: Länktips, vecka 3-4 | Public Good()

  • Evan

    I’m trying to understand why anyone resists this idea, other than because they perceive the requirement as somehow diminishing the status of business people, while elevating the status of regulators, and because they think business people deserve more valorization, and regulators less, they instinctively respond against.

    I think the idea is that these anti-fat activists keep claiming that these poor, poor people are being forced and tricked into eating fatty foods by the big corporations, and that if only they were better informed they would eat healthier. This study shows that, in fact, the reason people eat huge amounts of calories is because they want to. The corporations are simply doing their best to provide the delicious high calorie foods people want.

    Do people who eat fast food even know what calories mean? Or how many they burn in a day? Or how many calories adds up to a pound of fat?

    People at McDonalds are innumerate?! What a shock!

    I’m surprised there’s all this scorn being heaped at people who eat fast food. Stop being so elitist!

    I eat fast food all the time because it’s delicious, cheap, and saves time. I do make use of the calorie labels so I don’t get fat, but fast food really isn’t any less healthy than any other sort of food, as long as you eat it in moderation. I think the perception that it tastes bad and is unhealthy comes because it’s regarded as low-class, rather than because of the way it interacts with your taste buds or metabolism. If gourmet food was cheap people would pig out on that and get fat.

    When a food is advertised as low-calorie, I don’t think “healthy”, I think “small serving size”. To decide whether something is healthy, I look at the macronutrient ratios instead.

    Macronutrient ratios are good for making sure you get all the nutrients your body needs, but calories are more important if you’re trying to lose weight.

    • I was wondering whether the right information is being provided. What use are calorie counts if you don’t know (a) how many calories you burn every day and (c) that an extra 3500 calories makes about a pound of fat? I used to eat fast food, and while I’ve never been overweight, I definitely didn’t know the numeric specifics until like two years ago. I’m sure I’d heard it in mandatory high school health class, but I’d never paid attention.

  • dustydog

    Multiple trials showing the information makes a difference, one trial by partisans showing no difference, and you choose for your blog ___?

    Work harder on overcoming that bias, buddy.

  • I just added to this post.

  • I think this is the answer: posting calories increases calorie consumption because it increases cognitive load.

    This would be predicted by empirical studies like this one:

    Shiv, B. & Fedorikhin, A. (1999). Heart and mind in conflict: The interplay of affect and cognition in consumer decision making. Journal of Consumer Research, 26(2), 278-292.

    that showed that subjects tasked with remembering a 7-digit integer were more likely to choose unhealthy food than subjects told to remember a 2-digit integer.

    So, to get people to eat more calories, give them a bunch of three-digit integers to add!

    (The Baba Shiv study was publicized in a Radiolab episode.)