Food Subsidy Fails

Many developing countries use food-price subsidies or price controls to improve the nutrition of the poor. However, … consumers may then substitute towards foods with higher non-nutritional attributes (e.g., taste), but lower nutritional content per unit of currency. … We analyze data from a randomized program of large price subsidies for poor households in two provinces of China and find no evidence that the subsidies improved nutrition. (more)

This of course seems to be the median result for all randomized studies which try to improve people: no effect.

In the recent Fast Food episode of Penn & Teller’s Bullshit!, Brian Wansink of the Cornell Food & Brand Lab was shown separately asking two different groups to estimate the calories in a western chicken salad they had just eaten. Those told correctly that it was from Taco Bell correctly estimated its 970 calories, while those who were told it came from “California Garden Cafe; Gourmet garden-fresh cuisine” guessed about half as many calories.  Since one of main anti-fast-food proposals is for clearly-marked calorie counts on menus, this lab result suggests such proposals would hurt non-fast-food places more.  Anyone know how robust is this lab result, or if the proposals apply equally to all food sales?

GD Star Rating
Tagged as: ,
Trackback URL:
  • The fast food result is another example of the well-established Health Halo Effect. People who think a food is healthy tend to underestimate the calorie count.

  • Arthur B.

    The test group is not necessarily irrational, what if California Garden Cafe tends to produce similar tasting salad with lower calorie content ?

    In that case the proposal would hurt fast foods more.

  • Pingback: Tweets that mention Overcoming Bias : Food Subsidy Fails --

  • Jack (LW)

    Since one of main anti-fast-food proposals is for clearly-marked calorie counts on menus, this lab result suggests such proposals would hurt non-fast-food places more.

    Well, is it the case that food from places with names like “California Garden Cafe; Gourmet garden-fresh cuisine” sell salads with the same number of calories as the Taco Bell salad? My first reaction to the lab experiment is that giving a lower calorie estimate based on the healthy-sounding name probably is a good idea since it is pretty difficult to estimate calorie intake just by seeing and eating something. The experiment tells us non-fast food places could probably dupe people for a while, but it doesn’t tell us they really are.

    Fancy restaurants tend to have very high calorie meals though and the proposal would hurt them.

    • Jimmy

      As someone who actually likes calories, it’s not that tough to tell. Most people seem to think stuff like “fast food is bad and high calorie” then just assign a number that feels high. If you put in a tiny bit of effort (and check to calibrate yourself), you can come up with pretty decent guesses.

      Even better than the visual guess, the longer you go until you feel like you need to eat, the more calories it is. If you eat food that is greasier than normal, it’s hard to see the extra grease, but you’ll feel the extra calories.

  • If our food was just all organic, instead of making us pay a premium for it, I don’t think people would have half as many problems with their health and nutrition..

    • If organic food were cheaper people might prefer the non organic food.

      BTW I do not think that organic food is any healthier than other food.

    • J

      Why do you think that?

      • J

        To clarify, that question is for Beth.

    • Justin P

      “If our food was just all organic,” billions of people would die of starvation. Organic growing is more labour intensive and produces less food per acre than regular growing.

      “World hunger is projected to reach a historic high in 2009 with 1 020 million people going hungry every day, according to new estimates published by FAO today.”

      And wouldn’t you know it….most of those countries grow their food organically.

  • There are non-economic reasons why food subsidies fail, illustrated by the Penn & Teller experiment — calories don’t count, yet everyone is afraid of them.

    If you doubt, do a simple self-experiment to convince yourself. One week, eat 1000 calories per day consisting only of table sugar. Another week, eat 2000 calories per day consisting only of fatty meat — marbled steak, sausages, bacon, liver products, etc. Keep track of how vigorous you are and compare.

    Of course the contrast will be even more obvious if you carried out the diets for months rather than a week, but this will give you a good enough feel for why eating lots of calories is not unhealthy. Hell, the Ache hunter-gatherers of Paraguay eat between 3000 and 4000 calories a day, yet are lean and muscular. It’s because most of that comes from animals and low-carb plants like nuts and berries.

    • It’s because they are physically active. I have worked very physical jobs, less physical, and physically passive jobs; and without much change in diet over the decades, my weight closely tracks my activity level. I have tried, for extended periods (months to years) various vegetarian and low-carb type diets – food composition actually matters very little.

    • Look at Micheal Phelp’s diet. He is very lean but puts away a lot of all kinds of food including carbs.

      • Justin P

        If he isn’t in the water he is eating.

        They did a special on his diet during the Bejing Olympics…I can’t remember exactly what it all was…but lots of bacon, eggs, ham, milk…all that stuff that is “supposed” to kill you.

  • If you fear not getting enough calories on the all-sugar diet, then double both calorie counts: 2000 per day of sugar vs. 4000 per day of fatty meat.

    • magicdufflepud

      Your position might make sense if calories were magical little men who flew to your cells to make you feel like doing things. But they aren’t. They’re a measure of energy. And if you take in more energy than you expend, your body will store the excess, pro-rated for the additional energy cost of moving additional weight, as fat.

  • UserGoogol

    agnostic: I’d have to figure that a big factor is that the Ache hunter-gatherers of Paraguay are hunter-gatherers, and that requires a certain amount of exercise. Certainly different kinds of foods are metabolized differently and that effects weight and health, but to extrapolate from that really ignores the calories-out side of the equation.

    Calories in minus calories out equals calories stored. That is a basic consequence of conservation of energy. It’s more complicated than just calorie counting because how many calories you burn (basal metabolism as well as physical activity) and in what format (fat versus muscle) you store your excess calories is effected by a lot of things, including the non-caloric qualities of food. And of course, there’s more to nutrition than just weight.

  • What’s the motivation behind those posturing against a paternalistic requirement that food-sellers mark calories?

    You don’t want to see the calories?

    The economic cost seems slight to me, and although I may be a rent-seeking minority, I want to know the calories in all food products that I purchase, and I don’t want to do the relatively grueling work to look up and calculate and carry with me fuzzy best approximations of the numbers.

    • In other words, you want everybody to pay for information they don’t care about, because you want it and are lazy.

    • kevin

      The costs fall disproportionately on small operations, especially those that don’t serve a standardized menu. The cost for McDonald’s would be small, since they serve highly homogenized, standardized food, and the menu costs could be distributed across thousands of restaurants.

      Smaller restaurants often don’t have a formal recipe for each dish, so the calorie amounts would have to be estimated for each cook. If calories listings are not enforced they could just make up a guess, but this would be close to worthless. If calorie amounts are enforced, then small operations would have to homogenize their food.

      Tl;DR: Your rent seeking is going to bankrupt my favorite small ethnic restaurants.

      • Kevin,
        Sounds like a stretch to me and a stark example of epistemology warping lawyerly argumentation.

      • Entwhistle

        Just a note: the federal law going into effect applies only to chain restaurants with 20 or more outlets, precisely for the reasons you outline above. Every local calorie-labeling law I know about has a similar threshold. None of them apply to your favorite small ethnic restaurants, unless your favorite small ethnic restaurants are Chipotle and P.F.Chang’s.

  • diogenese

    Second Hopefully Anonymous.

    I just don’t see how anyone can be against calorie counts on menu. I STILL go to McDonalds — but its nice to easily incorporate calorie count into my decisions on what to eat while I’m there.

    What would be interesting is if a fast food chain looked at buying patterns before/after calorie counts prominently placed on the menu., Which I’m sure they already have internally.

  • I didn’t notice anyone here posturing against showing calories. Robin started out talking about subsidies, and suggested that people incorrectly estimate non-fast food categories. H.A and I had a discussion about calories on menus here. I am/was indifferent, I don’t think there is much cost or benefit (possibly there is for non-fast food, but I’d like to see an experiment similar to the one discussed in the link).

    Karl Smith (and some medical doctors”) think “calories in minus calories out” is a very poor way of analyzing things.

    • Proper Dave

      I’m sure there is also physicists that thinks conservation of momentum/energy etc. is a “poor way of analyzing” things…
      Let me guess fat is epiphenomenal?

      • What’s suggested is that there’s a lot that goes on inside the body that determines how much goes in and out. So some attempts at reducing calorie intake can result in the body trying to save more fat or causing much stronger feelings of hunger. Dr. Lustig thinks that insulin regulates this process.

  • Matt Flipago

    Why aren’t more people concerned that this would adversely affect all non chain, If you are a small restaurant, and even worse with a daily menu, how can you count your calories.

    Also this makes complete sense. If you only had 30 dollars a week to eat, would you go to a fast food restaurant. Absolutely not, you would eat large amount of rice and beans, and some other cheap veggies, because they are cheap and nutritional. If you were given 3000 dollars you would eat kobe beef and crazy deserts, so it’s quite reasonable to assume that food subsidies can make people less nutritious.

  • Ordinarily I love finding delicious food around,and always can’t help wanting to have some food.It becomes my hobby gradually.How many gourmet food can we taste during all the lifetime?Look forward.