A buyer recently asked us to confirm that our paprika oleoresin was "certified natural" for a pet food application. The question was reasonable and the answer was not what they expected: in the United States, there is no such category to be certified against. "Natural colour" is a marketing phrase. Under the animal food labelling regulations it has no legal meaning, and a supplier who issues a "natural certification" is certifying something that does not exist in the framework.
This matters commercially because pet food is a growing category for colour ingredients, and the labelling rules are not the same as for human food. A colour that is straightforward in a biscuit can be a labelling problem in a kibble. What follows covers the three provisions that decide the question, and what they mean when you are writing a purchase specification.
21 CFR 501.22 sets out what animal food labels must declare. Paragraph (a)(4) deals with colour, and its structure is the key to everything else: the regulation defines artificial color as including any colour additive, and it reaches that definition by reference to 21 CFR 70.3(f).
The consequence is a closed system with only two positions. A colouring substance is either a colour additive that appears in the listings, or it is not. There is no third category for materials that are natural, plant-derived, or traditionally used. Those descriptions may be true, and they may be useful in marketing, but they do not change which of the two positions the material occupies.
For a formulator this produces a specific practical question, and it is not "is it natural". The question is: which listing covers this material, in this article form, for this use? A paprika oleoresin and a paprika powder are separate entries in the United States, and quoting the oleoresin entry for a powder misstates the article. That distinction catches suppliers regularly, because both products are sold under similar names and both are described as natural.
It also explains why a "natural" claim on a pet food pack is a marketing decision rather than a compliance one. The compliance question is whether the colour used is covered by a listing; the word on the front of the pack does not affect that, and it cannot substitute for it.
Paragraph (a)(2) of the same section addresses a group of materials that sit across two categories: paprika, turmeric, saffron, and others that are used both for flavour and for colour. The rule is that these must be declared as spice and coloring unless they are declared by their common or usual name.
The practical meaning is narrower and more consequential than it first appears. The same drum of paprika can enter a pet food plant under two different declarations, and the choice is not cosmetic:
Which one is correct depends on why the material is in the formula and at what level. A seasoning-level addition reads as spice; a colour-level addition, where the material is present to deliver shade rather than flavour, is a different proposition. This is why the intended use has to be settled on the purchase order rather than inferred from the invoice description.
There is a second consequence that buyers discover later: the specification follows the declaration. A lot bought against a spice specification carries spice-trade microbiological limits and a flavour-driven quality standard. If it is then used at a colour-level dose, both of those assumptions fail — the microbial load is higher than the application can tolerate, and the colour value was never guaranteed in the first place. Paprika is the clearest example, but the same pattern applies to turmeric and to saffron.
The European approach asks a different question again, and it is the one that most often gets summarised incorrectly in supplier literature.
Feed additives in the EU are governed by Regulation (EC) No 1831/2003. Colourants fall within Annex I, Category 2 (sensory additives), point (a). Article 3(1) establishes the general principle: a feed additive may not be placed on the market or used unless it has been authorised under the Regulation.
The authorisation itself is species-specific. Article 7(3)(e) requires an application to specify the animal species and categories the additive is intended for, and Article 8(4)(c) requires the authorisation to state them. In other words, the unit of authorisation is not "paprika oleoresin" — it is "paprika oleoresin, for these species, at these conditions".
There is a simplification route. Article 7(5) provides a procedure for additives already authorised for use in food, intended to reduce the burden of demonstrating safety where the substance is already established. This is where the most common misconception sits.
The simplification procedure is not automatic mutual recognition. A food-grade authorisation does not transfer to feed use by operation of law. The applicant still has to make an application, still has to specify the species, and the outcome is still a feed authorisation with its own conditions. Treating a food approval as a feed approval is a mistake that surfaces at customs or at a customer audit, and by then the shipment is already moving.
For a buyer, the practical question in the EU is therefore narrower than in the US: not just which substance, but which species and which conditions, and can the supplier show the authorisation that covers them.
| United States | European Union | What customers often assume |
|---|---|---|
| The CFR entry covering the substance, and the article form it applies to | An authorisation under 1831/2003, naming the species and conditions | "A natural certification" |
| The correct declaration — spice and coloring, or the common name | Feed additive labelling requirements, including the species scope | "The same label as human food" |
| Batch documentation: COA, residual solvents, heavy metals | Batch documentation plus a statement of the authorised species range | "One COA is enough" |
One point of caution belongs here. The CFR does not set out a separate named provision for every Part 73 colour in animal food; the framework operates through the definition chain described above. That is a different statement from saying no listing is required, and the distinction matters — the applicable entry still has to be identified for the specific article, and that is a question for your regulatory team or your broker rather than for a supplier's product page.
Three things, in this order.
State the intended use. Colour or spice, and at what dose level. This single line determines which specification applies, which declaration is correct, and which microbiological limits are appropriate. Without it, the supplier is guessing and the certificate may be issued against the wrong standard.
Name the article form. Oleoresin and powder are different regulatory objects in the US and different products in a formula. A specification written for one does not cover the other.
Ask for the species scope, if the destination is the EU. Not "is it approved" but "which species and categories does the authorisation cover". A supplier who cannot answer that has not established the position for your market.
The labelling question in pet food is not really about whether a colour is natural. It is about which listing applies, how the material is declared, and — in Europe — which animals the authorisation covers. Those are answerable questions, and they are the ones worth asking before a formulation is fixed.
We supply colour materials with the documentation that these questions require, and we would rather set out the position per market up front than discover a labelling problem at the customer's audit. The natural colours range covers the grades discussed here; if you are specifying for pet food, tell us the species, the process window and the destination market and we will set out what each grade requires.
