Kilojoules, calories and kilocalories
One kilocalorie is exactly 4.184 kilojoules. The "calories" on a food label are kilocalories, so an 8,700 kJ daily reference is about 2,079 kcal. Dividing kilojoules by four is a useful estimate but runs roughly five percent high.
Updated 2026-08-25
The capital C nobody prints
A calorie, in the physics sense, is the energy needed to raise one gram of water by one degree Celsius. It is a tiny amount — so small that food would be labelled in hundreds of thousands of them, which is unusable.
So food labelling uses the kilocalorie: one thousand of those, sometimes written Calorie with a capital C to distinguish it. That convention has never worked, because nobody capitalises reliably and the two words sound identical when spoken. The result is that "a 500 calorie meal" means 500 kilocalories, or half a million calories in the strict sense, and everybody understands this without anyone ever saying it.
This is why nutrition labels in some countries print kcal rather than cal — it is the same number, stated unambiguously. If a label says 250 kcal and an app says 250 calories, they agree. If something claims a snack contains 250 calories in the strict small-c sense, it is wrong by a factor of a thousand, and you will almost never encounter that outside a physics classroom.
The practical rule: on anything to do with food, "calorie" means kilocalorie. There is no ambiguity to resolve in daily use, only a naming accident to be aware of when a number looks a thousand times off.
The exact factor, and the shortcut that is wrong
One kilocalorie is 4.184 kilojoules. That is a definition rather than a measurement — the thermochemical calorie was fixed at exactly 4.184 joules, so the conversion carries no uncertainty and never needs updating.
A second value, 4.1868, appears in older engineering references. That is the international steam table calorie, defined slightly differently, and it is not what food labelling uses. The difference is about a hundredth of a percent and will never matter for a meal, but it explains why two sources can disagree in the fourth digit without either being wrong.
The familiar shortcut is to divide kilojoules by four. It is genuinely useful for mental arithmetic and it is consistently wrong in the same direction: dividing by 4 rather than 4.184 overstates the calorie figure by about 4.6 percent. A 2,000 kJ meal is 478 kcal, and the shortcut says 500.
Five percent is small enough to ignore for a single snack and large enough to matter across a week. Someone eating 8,700 kJ a day and estimating by quarters believes they are eating about 2,175 kcal when the real figure is 2,079 — a gap of roughly a hundred kilocalories daily, which is the entire margin most people are trying to manage. Estimate by quarters to decide whether a sandwich is worth it; use the exact factor when the total is the thing you care about.
Why the unit depends on where you shop
Australia and New Zealand label food in kilojoules. Most of Europe and the UK print both kJ and kcal. The United States prints calories alone. All three are describing the same energy.
The kilojoule is the SI unit, which is why regulators that standardised later chose it — it belongs to the same system as metres and grams and needs no special-case definition. The calorie survived elsewhere because it was already in every cookbook, every diet book and every household, and changing an entrenched unit costs more in confusion than it gains in consistency.
The practical consequence shows up when the reference figures cross borders. Australian labelling uses 8,700 kJ as the average adult daily intake, and packets express their contents as a percentage of it. That figure is 2,079 kcal — deliberately close to, but not the same as, the 2,000 kcal reference used in the United States and the 2,000 kcal used on European labels. So the same biscuit can honestly be described as 5 percent of a daily intake in one country and 5.2 percent in another.
None of this changes what you eat. It does mean that comparing a percentage on an Australian packet with one on an American packet is comparing against slightly different denominators, and that an app set to the wrong region can quietly disagree with the label in your hand.
The number on the packet is already an estimate
It is worth knowing how the figure was produced before treating it as precise.
Most food energy values are not measured for each product. They are calculated using Atwater factors — fixed energy values per gram for each macronutrient, conventionally 4 kcal per gram of protein, 4 for carbohydrate, 9 for fat and 7 for alcohol. Multiply the composition by those and you have the label. Those factors are averages across food types, so the number is an estimate before it is ever rounded.
Then it is rounded, and the rounding rules differ by jurisdiction. A label may round to the nearest 10 kJ or the nearest kcal, and a value derived from an already-rounded number inherits that error. Converting a rounded kJ figure to kcal and back will not always return what you started with, which is not a fault in the conversion — the information was lost at the label.
Tolerances are wider than most people expect too. Regulators generally permit a substantial margin between the declared value and the measured one, because natural variation in ingredients is real and unavoidable. Two identical-looking portions of the same product can genuinely differ.
The useful conclusion is not that labels are unreliable but that precision beyond about three significant figures is imaginary. Converting 1,062 kJ to 253.8 kcal is arithmetically correct and falsely precise; the honest reading is "about 250". Chasing decimal places on a number that was estimated, rounded and legally permitted to vary is effort spent on the wrong part of the problem.