Fast Food Economics: How Medium Became the New Large

As a nutrition researcher tracking portion sizes and labels manufacturers use to describe such sizes, I have seen food portions not only grow larger over the years, but the labels to describe foods and drinks have also changed.

For example, when McDonald’s opened in the 1950s, the company offered one size soda, which was 7 ounces; today’s 12 ounces is labeled a kid’s size and the 16-ounce is labeled small. Similarly, when Burger King opened, the company offered a 12-ounce small and a 16-ounce large soda. The 12-ounce is no longer sold and the 16-ounce comes as part of the value meal. Burger King’s small soda is now 20 ounces, the medium is 30 ounces, and the large is 40 ounces.

Does anyone pay attention to these label descriptors? And do they influence how much we really eat? Apparently yes, according to a new study published in Health Economics by Cornell University researchers David Just and Brian Wansink.

The study found that labeling a food as “regular” or “double size” affects how much consumers will eat, regardless of how big or small the portion size actually is.

The researchers served subjects two different portions of pasta in either a one cup-portion or a two-cup portion. For some of the subjects, the two different size portions were labeled “half-size” and “regular.” For the other subjects, the identically-sized portions were labeled “regular” and “double-size.” The labels for the first group of subjects indicated that the two-cup pasta portion was the regular size, while it was suggested to the second group of subjects that the one-cup pasta portion was the regular size.

The study concluded that varying the “regular” portions affected how much the subjects actually ate. Subjects ate more food when the portion was labeled “regular” than when it was labeled “double-size” despite the fact that the two sizes were actually the same size.

The subjects were also willing to pay more for a larger sounding portion size.

As reported in newsLI.com, “These varying concepts of ‘regular’ portions made all the difference in how much people would spend and subsequently eat,” said Just. “Participants ate much more when their portion was labeled ‘regular’ than when it was labeled ‘double-size.’ In fact, participants who thought their portion was ‘double-size’ left 10 times the food on their plate.”

How does this study affect those of us who typically eat out at eateries that offer foods and drinks in different sizes? The chart below shows the sizes of fast food soda portions at top fast-food chains.


Kids 12 oz.
Small 16 oz.
Med 21 oz.
Large 32 oz.

Burger King

Value 16 oz.
Small 20 oz.
Medium 30 oz.
Large 40 oz.


Small 16 oz.
Medium 20 oz.
Large 30 oz.
Mega Jug 64 oz.

As you can see, the benign sounding “medium” soda is actually quite large. McDonald’s medium portion is 21 ounces (a pint and a half) and Burger King’s medium soda is 30 ounces (nearly a quart). But because these items are labeled medium, customers may consider themselves virtuous by not ordering the large, and may in fact order a medium order of fries to go with the soda.

My advice: Next time you visit an eating establishment that sells food in several sizes, I suggest ordering the small. Unless, you are visiting a Starbucks where the small is labeled tall.

Lisa Young is a nationally-recognized nutritionist. She is an adjunct professor at NYU and the author of The Portion Teller Plan. This post originally appeared in The Huffington Post.

9 replies »

  1. I am so glad I stumbled upon your site. I really found you by mistake, while I was browsing on Google for something else. Anyways I am here now and would just like to say thank you for a great post and an all round inspiring blog. (I also like the theme/design), I don’t have time to read through it all at the moment, but I have added your website to my favorites, so when I have time I will be back to read more. Please do keep up the awesome job!

  2. This variation in size labelling explains why my wife’s ordering of a medium soda (diet) yields such a wide range of results at different fast food establishments. Since the more significant implication is for those consuming sugar-sweetened beverages, there may be value to imposing federal uniform size standards to help ensure consumer understanding of what they’re purchasing. One would prefer that the fast food industry impose their own voluntary standards, but that hasn’t happened to date.

  3. Very creative approach. When my kid and I buy a soft drink (very rare occurrence), we get a medium, and we share it. If you take account of the ice, I suspect that he and I each end up with only about 8 oz of actual sugar-sweetened beverage.

    He’s only 9 but has known for some time (and been able to articulate) that soda is treat food and, if you have it, you do so infrequently and only after you’ve eaten your “good” food.

  4. Here’s an idea: choose a size of drink based on what you want to be. If you want to be large, drink large. If you want to be medium, drink medium. If you want to be small, drink small.

  5. Brilliant piece. If the right to vote were expanded to seven year old. Its policies would most definitely reflect the ‘legitimate concerns of children to have adequate and equal access to free’ french fries, lemonade and videos.

  6. I would warn against drawing any policy conclusions though. The Lucas Critique applies to these estimates–my guess is if you try to get people to eat less by requiring restaurants to re-label their food, we’ll just end up shifting consumers’ expectations, with no long-run change in behavior.

    That is, I don’t think the behavioral response to framing is necessarily irrational. It probably arises due to uncertainty over how much food consumers will find filling, for example. If “double” size consistently leaves them with less food, they will revise their expectations and start ordering “triple” size.