TempoLife › Features › Nutrition myths, checked › negative-calorie-foods
Some foods take more energy to digest than they contain, so eating them burns calories.
Digestion costs a fraction of a food's energy, not more than all of it. Even at the most generous figure ever measured for a whole food, the classic "negative calorie" list still leaves you ahead.
Verdict: Falsenegative calorie foodsthermic effectTEF
The claim, as people state it
“Some foods take more energy to digest than they contain, so eating them burns calories.”
What is actually known
The cost of digesting, absorbing and storing food is called the thermic effect of food, and it is well measured. Across a mixed diet it runs at roughly a tenth of intake. It varies by macronutrient: protein is the expensive one, at roughly 20-30% of the calories it carries; carbohydrate sits near 5-10%; fat is the cheapest, near 0-3%. Nothing exceeds 100%, which is what a genuinely negative-calorie food would require.
The foods on every version of the list — celery, cucumber, lettuce, grapefruit, watermelon — are watery, fibrous and carbohydrate-based. They are the cheapest kind of food to digest in absolute terms because there is barely any energy in them to process, and their thermic effect sits at the low end of the range, not above 100%.
The one place the arithmetic gets interesting is fibre, some of which passes undigested and yields less energy than a food label credits it with. That is a genuine and known correction of a few calories, not a reversal of sign.
It is worth being clear about where the thermic effect actually comes from, because it is not chewing. It is the metabolic work of breaking food down, transporting nutrients across the gut wall, and converting or storing what arrives — protein is expensive because rebuilding amino acids into human protein, or converting them for fuel, costs energy that fat storage does not. That is why the ranking runs protein, then carbohydrate, then fat, and why no arrangement of it reaches the far side of zero.
Why the myth persists
Because the foods on the list are genuinely excellent for anyone eating less: enormous volume, negligible energy, real fibre. The advice that comes attached to the myth — eat more of these — is good advice. Only the explanation is wrong, and a good outcome protects a bad explanation for a long time.
The arithmetic
Below is the classic negative-calorie list, priced from the food database and charged the most generous digestion cost ever measured for a whole food — 30%, the top of the range for pure protein. These foods are not protein, so their real cost is lower than this. Even at 30%, every row stays positive.
net kcal = kcal per 100 g − (kcal per 100 g × 0.30)
| Food | kcal / 100 g | Digestion cost at 30% | Net kcal / 100 g | Net in a 300 g serving |
|---|---|---|---|---|
| Celery | 14 | 4.2 | 9.8 | 29 |
| Cucumber | 15 | 4.5 | 10.5 | 32 |
| Lettuce | 17 | 5.1 | 11.9 | 36 |
| Watermelon | 30 | 9.0 | 21.0 | 63 |
| Grapefruit | 42 | 12.6 | 29.4 | 88 |
| Apple | 52 | 15.6 | 36.4 | 109 |
| Broccoli | 34 | 10.2 | 23.8 | 71 |
For a food to be negative-calorie, the last two columns would have to be below zero, which requires a digestion cost above 100%. The highest figure ever measured for any whole food is roughly a third of that, and it belongs to protein rather than to celery.
Every figure in this section was computed when this page loaded, from 5 237 food rows and 31 exercise rows already in this repository. Nothing in it is stored as prose.
The honest nuance
The practical point survives the myth intact. Filling half a plate with these foods genuinely reduces the calories in the meal, because they take up the space something denser would have occupied. That is displacement, and it works.
Chewing burns energy too. It is measurable, and it is small enough that no realistic amount of chewing changes the answer.
General information, not a diagnosis and not medical advice. Anything touching a condition you have, a medicine you take or a test result belongs with your own clinician, who has information no web page does.
Frequently asked questions
Is there any food with a negative net energy?
No food yet measured. Water and most non-caloric drinks are zero, and ice water costs a trivial amount to warm, but that is not the same claim.
What about drinking cold water?
Warming a litre of ice water to body temperature costs on the order of 30-40 kcal. It is real, it is small, and it is the closest thing to a negative-calorie effect that exists.
Source: USDA FoodData Central (SR Legacy) via the TempoLife food database; MET values from the 2011 Compendium of Physical Activities · checked 2026-09-02
Work it out for yourself
Other claims checked
Eating after 8pm makes you gain weight.
The clock does not add calories. What the clock does is change how much you eat and how well you sleep — and those are real…
Breakfast is the most important meal of the day — skip it and you will gain weight.
Breakfast eaters do tend to be leaner. Adding breakfast to someone who was skipping it does not reliably make them leaner. Those…
Detox teas flush toxins out of your body and help you lose weight.
No product on this market names the toxin it removes or measures it before and after. The weight that moves is water and gut…
Do enough crunches and the fat over your abs will go.
Training a muscle builds that muscle. It does not preferentially withdraw fat from the skin above it. Where fat leaves from is…
The more you sweat, the more fat you are burning.
Sweat is a cooling system, not an exhaust pipe. It measures how hot you are, not how hard you worked — and every gram of it comes…
Do not eat carbohydrates after 6pm.
The same slice of bread has the same energy at 18:01 as at 17:59. This rule works, when it works, by deleting a category of food…
All claims · Weekly research digest
Stop guessing at your own numbers
Almost every myth on this site ends at "it depends on your total intake and your total movement". TempoLife records both, so the answer stops being a debate.