US & Europe · Independent Updated August 2026
The Banned in Europe Registry
Living Differently

Portion Sizes: America vs Europe, Actually Measured

Researchers weighed the food. A Philadelphia plate ran about 25 percent heavier than the same dish in Paris. The interesting part is what that gap does and doesn't explain.

By 8 min read Living Differently

You come back from ten days in Italy and your own kitchen looks strange. The bowls seem enormous. The coffee cup you’ve used for years now reads as a soup vessel. You didn’t count a single calorie abroad, you ate bread and cheese and gelato, and somehow your jeans fit better.

Everybody has a theory about why. Most of those theories go straight to chemistry: the additives, the seed oils, the bread. And some of that is worth taking seriously, which is most of what this site is about. But before we get to what’s in the food, there’s a duller and much better documented variable sitting right in front of us.

How much food is on the plate.

Somebody actually weighed it

In 2003, a team led by Paul Rozin at the University of Pennsylvania, working with Claude Fischler in France, did the unglamorous thing. They walked into eleven matched pairs of restaurants in Paris and Philadelphia, fast food places, pizzerias, ice cream shops, a Chinese restaurant, a local diner, and they weighed the portions.

Mean portion size in Paris: 277 grams. In Philadelphia: 346 grams. That’s about 25 percent heavier on the American side, for what was nominally the same dish. The Chinese restaurants were the extreme case, with Philadelphia servings running roughly 72 percent larger than their Paris counterparts. The researchers also found the pattern held outside restaurants: individual portions in supermarkets, portions specified in cookbooks, and the sheer number of “all you can eat” listings in dining guides all skewed larger in the US.

And then the detail that sticks with me. They timed people eating at McDonald’s in both cities. The French took longer. Smaller plate, more time with it.

The paper was called “The Ecology of Eating,” and the argument was that the environment does a lot of the work we usually attribute to willpower or genetics. You eat what’s in front of you. Make what’s in front of you 25 percent bigger, and you eat 25 percent more, without ever making a decision about it.

The American portion didn’t arrive fully formed

It grew. Lisa Young and Marion Nestle at NYU tracked marketplace portion sizes against federal standard serving sizes and found that most portions on offer exceed the standard serving by at least a factor of two, and in some cases by eight. Fast food portions in particular ran two to five times the size of the original versions those chains launched with.

The timeline matters. Portions started creeping up in the 1970s, jumped sharply in the 1980s, and kept climbing in parallel with average body weight. That’s a correlation, not a smoking gun, and Young and Nestle were careful about saying so. But it’s a correlation with a plausible mechanism and it’s held for fifty years.

A separate 30-year audit, published in the Journal of the Academy of Nutrition and Dietetics, compared menus at ten American fast food chains in 1986, 1991 and 2016. Entrées got bigger. Desserts got much bigger. Entrée calories rose roughly 30 per decade, desserts roughly 62, and most of that increase came from mass rather than from any change in recipe. The food didn’t get denser. There was just more of it.

Meanwhile the number of items on those menus grew 226 percent. More choices, each one larger.

Why the plate wins

The reason this matters more than it should is a quirk of how people decide they’ve had enough.

Rozin’s group, working with Andrew Geier, later described something they called unit bias: the tendency to treat one unit of a food, one cookie, one bagel, one serving, as the correct amount, regardless of how big that unit happens to be. Double the size of the unit and people don’t eat half of it. They eat the unit.

That’s a devastating little finding when you set it next to a menu that has quietly doubled its units over forty years. It means the plate is doing your portion control for you, and it has been doing it worse every decade.

It also explains why the fix that feels obvious, just eat less, is so much harder in one country than the other. You are not fighting your appetite. You’re fighting a default. And defaults are extremely good at winning.

The label tells you which country you’re in

Here’s the part that gets missed, and it’s the closest thing to a regulatory story in all this.

In the EU, under Regulation (EU) No 1169/2011, the nutrition declaration on prepacked food is mandatory and it must be expressed per 100 grams or per 100 millilitres. Per portion is allowed as an extra, never as a substitute. So every European label forces the same denominator. A crisp packet and a family-size crisp packet report the same number, and you can compare across brands without doing arithmetic.

In the US, the Nutrition Facts panel is anchored to the RACC, the Reference Amount Customarily Consumed. The RACC is supposed to reflect what people actually eat in one sitting, which sounds sensible until you notice which direction it moves. In the 2016 final rule updating serving sizes, the FDA raised the RACC for carbonated beverages from 8 fluid ounces to 12, and defined a 20-ounce soda bottle as a single-serving container, because nobody was splitting one. The rule was right about behavior. It was also, in effect, the government’s own portion size ratcheting up to meet the market’s.

Two different philosophies sitting in the fine print. One label asks what’s in a fixed unit of this food. The other asks how much of this food a person like you tends to put away.

What this does not explain

I want to be careful here, because the portion story is seductive and it is not the whole answer.

American adult obesity was 40.3 percent in the CDC’s NHANES data for August 2021 through August 2023, with another 31.7 percent overweight. That’s roughly 72 percent of American adults above a healthy BMI, and those are measured heights and weights, not self-reported ones. Eurostat’s 2022 figures put 50.6 percent of EU adults in the overweight-or-obese category, with obesity rates by country running from about 6 percent (women in Italy) to nearly 29 percent (men in Malta).

Treat that comparison gently. The European numbers come from EU-SILC, which asks people to report their own height and weight, and people reliably shade both in the flattering direction. So the true European figure is higher than 50.6 and the gap is narrower than the raw numbers suggest. It is still a gap, and it is still much too large to hand to portion size alone.

Plenty of other things differ. Europeans walk more, and not as exercise, just as transit. Meals are more likely to be sit-down events with a start and an end rather than a continuous grazing field. Snacking norms are different. Sleep, work hours, car dependency, the price of a soda relative to a piece of fruit, all of it stacks.

And portion size is also downstream of other things. It’s cheap to add volume to a plate when the calories are cheap, and American agricultural policy has spent decades making them cheap. So “bigger portions” is partly an answer and partly a symptom.

The additive story and the portion story are not competitors, either. A US-formulated cereal may carry colorings its European twin doesn’t, and the American box may pour a larger bowl. Both can be true. Our US vs UK label comparisons keep finding exactly that: different ingredients and different serving assumptions, side by side.

What to actually do with this

Nothing dramatic. A few things that follow directly from the evidence:

Read the per-100g column when it exists, and calculate it when it doesn’t. It’s the only number that lets you compare two products honestly, and it’s the reason European labels feel less slippery.

Watch the container, not the serving line. The FDA’s own rule concedes that a 20-ounce bottle is one serving because that’s how people drink it. Your snack bag is probably the same. The number on the panel is a description of a habit, not a prescription.

Notice the ecology. Rozin’s point wasn’t that the French have better self-control. It’s that they were handed less food and given more time to eat it, and the outcome followed. You can rebuild a small piece of that at home with smaller plates, one plating instead of a serving bowl on the table, and a meal that has an actual ending.

And skip the guilt. The most useful finding in this whole literature is that the environment was doing the deciding, not you.

Sources