How to Calculate the Carbohydrate in Pet Food (NFE): A Worked Guide
Carbohydrate is one figure you will almost never find printed on a bag of pet food: neither Regulation (EC) No 767/2009 in the United Kingdom and European Union, nor AAFCO rules in the United States, require it (FEDIAF, 2019; AAFCO, 2024). The accepted way to estimate it is by difference, through nitrogen-free extract (NFE), a calculation recognised by AAFCO and the NRC as the most accessible method short of a lab assay (NRC, 2006). This guide walks through the formula step by step, with figures you can recompute, then shows how to read the result for a cat and for a dog, and where the estimate stops being reliable.
Last updated :General documentary information. For an individual animal, a veterinarian's advice takes precedence over any online content.
Why isn't carbohydrate printed on the label?
Answer capsule. Because no regulation requires it. Regulation (EC) No 767/2009 in the EU and UK and AAFCO rules in the US list protein, fat, fibre and ash among the mandatory constituents, but not carbohydrate (FEDIAF, 2019). Since it is worked out by difference from the others, lawmakers left it optional, so the buyer must estimate or request it.
The absence follows from the regulatory frame, not from any wish to conceal. The four base analytical constituents, crude protein, crude fat, crude fibre and crude ash, are mandatory, with moisture required above a threshold; carbohydrate is simply deduced from these (Regulation (EC) No 767/2009). The mandatory block therefore supplies, indirectly, everything you need to estimate it.
The catch is that carbohydrate is sometimes the largest single fraction of a recipe. It can make up more than 40% of a kibble's dry matter, a major component that never appears on the pack (Tufts Petfoodology, 2021). For an obligate-carnivore cat, or a diabetic animal, this invisible figure is precisely the one that matters most, which is why learning to estimate it is worth the few minutes it takes.
What is the NFE formula, step by step?
Answer capsule. NFE = 100 - moisture - crude protein - crude fat - crude fibre - crude ash. The result estimates digestible carbohydrate, mainly starch and sugars. Worked example for a kibble at 8% moisture, 32% protein, 18% fat, 3% fibre and 7% ash: 100 - 8 - 32 - 18 - 3 - 7 = 32% carbohydrate as-fed (NRC, 2006; AAFCO, 2024).
Nitrogen-free extract is a remainder: rather than assaying starch directly, it subtracts every measured constituent from 100, and what is left is taken as digestible carbohydrate (NRC, 2006). The remainder gathers mainly starch and digestible sugars, plus a share of soluble fibre that the crude fibre method does not capture.
Work it through slowly with the example. Start at 100. Subtract moisture and protein: 100 - 8 - 32 = 60. Subtract fat and fibre: 60 - 18 - 3 = 39. Subtract ash: 39 - 7 = 32. The kibble holds an estimated 32% carbohydrate as-fed. The one prerequisite is that you have all five constituents, including ash and moisture; if the label omits either, the calculation becomes uncertain and the measured value is better requested from the maker.
How do you convert the result to a dry-matter basis?
Answer capsule. Divide the as-fed figure by the dry matter (100 - moisture) and multiply by 100. For the example kibble at 8% moisture, dry matter is 92%, so 32% as-fed becomes (32 / 92) x 100, about 34.8% carbohydrate on a dry-matter basis (AAFCO, 2024). The conversion lets you compare products of different moisture fairly.
The as-fed figure is useful, but it carries water, so it is not comparable across formats. Converting to a dry-matter basis removes the water and expresses the carbohydrate on the genuinely nutritive fraction. The single formula is: dry-matter rate = (as-fed rate divided by (100 minus moisture)) times 100, identical to the conversion you would use for protein or fat.
A second worked case shows the pattern holds. For a kibble at 10% moisture, 30% protein, 12% fat, 3% fibre and 7% ash, NFE is 100 - 10 - 30 - 12 - 3 - 7 = 38% as-fed, and on a dry-matter basis (38 / 90) x 100, about 42.2% (Tufts Petfoodology, 2021). That is more than 40% of the dry matter as carbohydrate, a substantial share for a food a buyer might have assumed was protein-led.
How reliable is a by-difference figure?
Answer capsule. It is an estimate, not a measurement. Because NFE is a remainder, it accumulates every imprecision in the other assays and folds in soluble fibre, so it tends to slightly overestimate truly digestible carbohydrate (Tufts Petfoodology, 2021). It gives a reliable order of magnitude for comparing foods, but a lab assay is preferable for a carbohydrate-sensitive animal.
The method is sound for what it is, and recognised by AAFCO and the NRC as the standard accessible approach (NRC, 2006; AAFCO, 2024). Its reliability depends on two things: the quality of the underlying measurements and the completeness of the label. Any error on protein, fat, fibre, ash or moisture carries straight through to the result, and a label missing ash makes the estimate frankly uncertain.
The practical rule is to match the tool to the stake. To compare two foods and spot a trend, NFE on a dry-matter basis is enough. To manage the diet of a diabetic cat or an overweight animal, where carbohydrate control is clinically important, the lab-measured value, supplied by a serious maker on request, removes the doubt (WSAVA, 2021). The estimate orients the choice; the measurement settles it. A useful habit is to record the as-fed inputs you used, so that if a maker later supplies a measured value you can see how far the by-difference figure strayed and recalibrate your reading of similar labels.
What carbohydrate level is high for a cat?
Answer capsule. There is no legal threshold, but many practitioners use a cautionary benchmark around 20 to 25% carbohydrate on a dry-matter basis for the cat, an obligate carnivore (NRC, 2006). Many cat kibbles exceed it, reaching 30 to 40% or more, because starch is needed for extrusion (Tufts Petfoodology, 2021). The figure is read on a dry-matter basis, never as-fed.
The cat gains little from carbohydrate: metabolically it is adapted to a diet rich in animal protein and low in starch (NRC, 2006). No regulation sets a maximum, but a frequently cited indicative benchmark sits around 20 to 25% on a dry-matter basis, estimated via NFE. Above that, the carbohydrate share moves beyond the species' physiological need.
Cat kibbles often exceed this because the manufacturing process requires it: a share of starch gives the kibble cohesion and expansion, so 30 to 40% carbohydrate on a dry-matter basis is common, and "grain-free" versions do not mechanically lower it (Tufts Petfoodology, 2021). The level becomes a genuine concern mainly for the diabetic, overweight or predisposed cat, where a higher-protein, lower-carbohydrate food, often a wet food, may be advised on veterinary guidance (WSAVA, 2021). For a healthy, stable-weight cat, weight monitoring comes first.
What level is acceptable for a dog?
Answer capsule. The dog tolerates carbohydrate better than the cat and has no regulatory threshold. A dog kibble commonly holds 30 to 50% carbohydrate on a dry-matter basis, acceptable for a healthy, active animal (NRC, 2006). An overweight or diabetic dog warrants a more measured share, to confirm with a vet.
The dog is an omnivore with a carnivorous lean and digests carbohydrate more efficiently than the cat. It produces the pancreatic amylase needed and carries genetic adaptations for starch digestion that the strictly carnivorous cat lacks (NRC, 2006). This is the physiological reason the benchmarks differ between the two species, even though excess carbohydrate still adds to energy density in the dog.
A level of 30 to 50% on a dry-matter basis is compatible with good health in an active, stable-weight dog. Below 30% usually signals a more meat-based recipe; above 50% is worth watching, particularly if the animal is overweight or diabetic (WSAVA, 2021). For a healthy dog, the exact figure matters less than the overall balance of the ration and a fit weight tracked by body-condition score. Reading carbohydrate alongside the energy density, where available, gives a fuller picture than the starch share alone, since two foods at the same carbohydrate level can still differ in calories.
Cat vs dog: reading the carbohydrate figure side by side
The table sets indicative carbohydrate benchmarks (dry-matter basis, estimated via NFE) for each species, with the context that shapes them. There is no legal threshold for either.
| Carbohydrate (DM basis) | Cat (obligate carnivore) | Dog (omnivore, carnivorous lean) |
|---|---|---|
| Below 20% | Close to the feline need | More meat-based recipe |
| 20 to 25% | Cautionary benchmark | Comfortably tolerated |
| 30 to 40% | High; common in extruded kibble | Within the common range |
| Above 50% | Very high; vigilance for at-risk cats | To watch if overweight or diabetic |
| Clinical context | Diabetes, excess weight raise concern | Vet advice if disease present |
The table shows the asymmetry: a level that is unremarkable for a healthy dog can sit well above what suits an obligate-carnivore cat. Read the figure against the species first, then the individual animal.
A clear method for estimating and reading carbohydrate
The recommendation is a short, repeatable routine. First, gather all five constituents from the label, protein, fat, fibre, ash and moisture, and apply NFE = 100 minus their sum to get carbohydrate as-fed (NRC, 2006). If ash or moisture is missing, stop and request the measured value rather than guessing. Second, convert to a dry-matter basis with (rate divided by (100 minus moisture)) times 100, so the figure is comparable across products (AAFCO, 2024).
Third, read the converted figure against the species. For a cat, treat 20 to 25% on a dry-matter basis as a cautionary benchmark and expect many kibbles to exceed it; for a dog, 30 to 50% is generally acceptable in a healthy animal (NRC, 2006; Tufts Petfoodology, 2021). Fourth, weigh the clinical context: a diabetic or overweight animal is where a high carbohydrate share genuinely warrants attention and a veterinary conversation (WSAVA, 2021).
Finally, keep the estimate in proportion. NFE is a reliable order of magnitude for comparing foods, not a lab figure, and it slightly overstates digestible carbohydrate by including soluble fibre (Tufts Petfoodology, 2021). Use it to orient a choice, request a measured value when an animal's health depends on it, and never read carbohydrate as-fed, which would flatter a wet food and distort any comparison.
---
Keep reading
Related questions: How do you work out the carbohydrate of a kibble that doesn't print it? | What is the carbohydrate-by-difference (NFE) formula? | At what level is a kibble's carbohydrate considered high for a cat?
Glossary: nitrogen-free extract (NFE) | dry matter
Hub: Reading and decoding a label
Sources: NRC, Nutrient Requirements of Dogs and Cats (2006); AAFCO, Understanding Pet Food and Calorie Content (2024); Regulation (EC) No 767/2009 (EUR-Lex; retained in UK law); FEDIAF Code of Good Labelling Practice (2019); WSAVA Global Nutrition Guidelines (2021); Tufts Petfoodology, Carb Confusion (2021).