TL;DR: Under [EU Cosmetics Regulation 1223/2009](https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX:32009R1223), brightening products sit firmly in the cosmetic category
TL;DR: Hydroquinone, which was historically used as a skin-lightening agent, lost its tentative final monograph status in 2020 and is now regulated as a new drug requiring full NDA submission
Key Technical Parameters #
Getting a brightening or whitening product registered across three or four markets simultaneously is genuinely one of the more complex compliance exercises in cosmetics. The challenge isn’t any single regulation — it’s that the EU, US, China, and key ASEAN markets each treat “whitening” through a completely different regulatory lens, and an ingredient that clears one framework can trigger a mandatory registration process in another. Brand developers focused on brightening and whitening SKU development tend to hit the same walls: they scope a formula for one market, then discover mid-stability that a second market requires a different classification or different evidence package. We see this pattern in most multi-market briefs that come through our lab. The key technical insight is this: the compliance burden is driven less by the active itself and more by the claim language and product category classification — and those two things are decided before a single gram of raw material is ordered.
How Each Market Classifies Brightening Products — and Why It Changes Everything Downstream #
The single most consequential decision in a multi-market brightening project isn’t the active concentration. It’s whether your product gets classified as a cosmetic, a quasi-drug, or an OTC drug in each target market. That classification determines your evidence package, your label restrictions, your registration pathway, and in some markets, whether you can launch at all without a pre-market approval.
Under EU Cosmetics Regulation 1223/2009, brightening products sit firmly in the cosmetic category. There is no quasi-drug category in the EU. This means you work within Annexes II, III, and V — the restricted, conditionally permitted, and preserved substance lists. Alpha-arbutin, for example, is currently permitted up to 2% in face products and 0.5% in body lotions under Annex III. Kojic acid is under SCCS Scientific Opinion review and has been restricted to 1% in face wash-off products and 0.7% in face leave-on products — with body application still contested. These aren’t guidance values. They’re hard limits. Exceeding them means your product is non-compliant and cannot be placed on the EU market regardless of your safety data.
In the US, the FDA does not recognize “whitening” or “brightening” as OTC drug claims for skin. The FDA OTC Drug monograph system — the framework that governs products like sunscreens and acne treatments — has no approved monograph for skin-lightening actives. Hydroquinone, which was historically used as a skin-lightening agent, lost its tentative final monograph status in 2020 and is now regulated as a new drug requiring full NDA submission. For practical purposes, this means US brightening products must be positioned and labeled as cosmetics. Claims like “reduces the appearance of dark spots” are acceptable cosmetic claims; “lightens skin” starts to push toward drug territory and should be avoided unless you want FDA scrutiny. The FDA Cosmetics Guidelines framework draws this line through claim language, not ingredient identity.
China is the most demanding market for this category. Under NMPA’s cosmetic regulation framework, products making whitening claims (美白) fall into the special-use cosmetic category, which requires pre-market registration rather than simple notification. As of 2021, the updated Cosmetic Supervision and Administration Regulation (CSAR) and its implementing provisions define whitening cosmetics as those claiming to “reduce melanin deposition and lighten skin color.” This triggers a full dossier submission including efficacy data, safety assessment, and in some cases human trial results. The registration process through NMPA Cosmetic Regulation typically takes 9 to 15 months for foreign brands, sometimes longer. Brands that try to avoid the special-use classification by softening claim language — using phrases like “brightening” or “luminosity” instead of “whitening” — sometimes succeed, but NMPA has been tightening its interpretation of borderline claims since 2022, and we see rejection notices for this strategy more often now than three years ago.
The table below captures the core classification differences that drive downstream compliance decisions:
| Market | Classification Category | Pre-market Approval Required? | Key Restricted Actives | Evidence Standard |
|---|---|---|---|---|
| EU | Cosmetic only | No (but CPNP notification required) | Alpha-arbutin ≤2% (face), Kojic acid ≤1% (face wash-off) | Safety Assessment per Annex I |
| USA | Cosmetic (no OTC whitening monograph) | No | Hydroquinone banned as OTC | Claim-based risk — no mandatory efficacy data |
| China (NMPA) | Special-use cosmetic (if whitening claim) | Yes — pre-market registration | Hydroquinone prohibited; whitening actives require NMPA-approved status | Efficacy + safety dossier, human trial data often required |
| Japan | Quasi-drug | Yes — MHLW notification | Arbutin ≤7% (quasi-drug), kojic acid permitted under quasi-drug framework | In vitro or clinical evidence required |
| ASEAN (via AEC) | Cosmetic | No (post-market notification in most member states) | Follows ASEAN Cosmetic Directive, aligned with EU Annexes in most categories | GMP compliance + safety data |
Japan deserves a note here because it’s the outlier that surprises brands most. The quasi-drug category is actually advantageous for brightening — it allows you to make explicit efficacy claims about melanin suppression that would be off-limits in EU or US cosmetic positioning. But the trade-off is MHLW notification, Japanese-language labeling, and an approved active list that is both more permissive and more specific than what EU or NMPA allow. Arbutin at 7% in a quasi-drug leave-on is legally compliant in Japan. That same formula at 7% would exceed EU limits by more than three times.
The Compliance Layer Nobody Briefs Us On: Test Method Standards #
Classification and ingredient limits get most of the attention in compliance discussions. But the test method standards — the technical frameworks that determine how you generate and present your evidence — cause at least as many launch delays in our experience.
ISO 16128 is the one that comes up constantly in European briefs right now. It’s a voluntary standard, not a regulatory mandate, but the market pressure from EU retailers and natural beauty positioning has made it functionally unavoidable for brands targeting that segment. ISO 16128 defines a natural and organic index for cosmetic ingredients, running from 0 to 1. The calculation methodology sounds straightforward until you actually apply it to a brightening formula. Many high-performing brightening actives — including most synthetic derivatives of natural compounds — score low on the natural index despite being well-tolerated and safe. Ascorbyl glucoside, for example, is produced via enzymatic synthesis; its ISO 16128 index score depends on how your supplier characterizes the manufacturing route. We’ve had batches of the same active from two different suppliers calculate to different natural index scores because of differing interpretations of the standard’s derivation rules. Voluntary doesn’t mean simple.
ISO 11930 is the preservative efficacy test that underpins the EU’s requirements under Annex I of EU Cosmetics Regulation 1223/2009. For brightening formulas, this matters more than it might in a standard moisturizer because the pH range that most brightening actives require — typically 4.0 to 5.5 for vitamin C derivatives, lower for L-ascorbic acid — creates a preservation environment that conventional challenge testing doesn’t always reflect accurately. A formula that passes ISO 11930 at pH 5.0 may have a very different microbial profile if the consumer’s usage pattern shifts pH toward 6.0 over time. We flag this in every brightening brief that comes with a low-preservation-load request.
If the formula carries any SPF claim alongside brightening, ISO 24444 becomes mandatory for EU market compliance. In vitro methods have been gaining acceptance for label SPF, but the EU still requires in vivo ISO 24444 testing for primary SPF claims. This adds 8 to 12 weeks and meaningful per-sample cost to the evidence package — and it needs to be planned at the formula-lock stage, not after stability is complete.
PCPC/INCI naming is a compliance layer that causes surprisingly frequent problems at the customs declaration and labeling stage. The PCPC Guidelines set the standard for INCI nomenclature used in the US and referenced internationally. China uses its own INCI-equivalent list (the IECIC, International Existing Cosmetic Ingredient Catalogue), and there are ingredients with established INCI names that appear under a different designation or with a restricted status in China. Brands that copy-paste their EU or US ingredient declarations onto Chinese labels without cross-referencing the IECIC have had product detained at customs. This is not a theoretical risk — our quality documentation team flags this as a real project delay we have encountered.
What Gets Misdiagnosed: The Claim Language Problem #
This is the root cause that generates the most project restarts in this category. A brand writes a brief for a brightening serum, the formula is developed and stabilized, and then — at label review stage — someone flags that the intended claim language triggers a different regulatory classification than originally planned. By that point, the formula is locked, the packaging is in production, and the compliance documentation has been built for the wrong evidence standard.
The mechanism works like this. A claim that reads “visibly brightens and evens skin tone in 4 weeks” is a cosmetic claim in the EU and US — it describes an aesthetic change to appearance. But add “by inhibiting melanin production” and you’ve crossed into functional mechanism language that, in some markets, either requires substantiation data at a level cosmetic brands don’t typically hold, or triggers product reclassification. In China, any reference to melanin, tyrosinase, or melanocyte activity in claim language — even indirect references — is read by NMPA as a whitening claim requiring special-use registration. We’ve seen briefs where the brand’s marketing team wrote the claim copy independently from the regulatory team, and the two didn’t converge until label design was 60% complete.
The threshold is not intuitive. “For a more radiant complexion” clears every market we work with. “Reduces the appearance of dark spots” is fine in the US and EU, requires careful framing in China. “Whitens skin” requires pre-market registration in China, positions the product in quasi-drug territory in Japan, and is functionally off-limits in the US for any cosmetic product. A three-word difference in claim language can mean a nine-month registration timeline difference between markets.
From a clinical evidence standpoint, the substantiation requirements also differ by market and claim type. A 2019 randomized, split-face controlled trial (n=44, 16 weeks) evaluating a 2% alpha-arbutin formula against vehicle control showed a 24.7% reduction in ITA angle variance (measured by spectrophotometric colorimetry) — the kind of data that supports a meaningful “evens skin tone” claim in the EU. That same data would not satisfy NMPA’s special-use registration requirements, which have shifted toward preferring Chinese population-specific panels and domestic CRO data. Two studies with the same methodology and similar endpoints can have very different regulatory value depending on which market you’re filing for.
Honestly, this is where brands lose the most time. The clinical evidence and the claim language need to be designed together, from the start, with target market compliance in mind. Retrofitting one to the other is expensive.
Prevention: What to Lock Before You Brief a Formulator #
This section is short because it should be. The information you need to specify before formulation begins is not complicated, but it is decisive.
First: list your target markets in priority order with intended launch sequence. Not “global” — specific countries. China registration timelines are 9+ months. If China is market three, that’s not a blocking constraint for launch. If it’s market one, it changes what claims you can develop, what actives you can use, and what your evidence package needs to look like.
Second: lock your claim language in draft form before any formulation begins. Not final — draft. Enough to determine which regulatory classification applies in each market. Run it past a regulatory consultant or your OEM’s compliance team before the formula is scoped.
Third: request your supplier’s IECIC cross-reference and EU Annex III compliance confirmation for every brightening active at the concentration you intend to use. This is a document you should be holding before the formula moves to stability.
What we ask for in our intake process is captured in our internal QC-07 Regulatory Pre-Screening checklist — it’s a one-page form, but filling it out forces the brand team to answer the questions that prevent downstream restarts. Ask your OEM if they have an equivalent.
Formulation Notes for Brand Partners #
When you brief us on a brightening or whitening project, the first questions we ask are: which markets, in what sequence, and what claim language does your marketing team currently have in mind? Those three questions determine more about your formulation parameters than any active selection decision.
The most common brief mistake we see is framing the regulatory question as “what’s the highest concentration we can use?” The better question is “what concentration clears all target markets with the claim language we want?” Those two questions often have different answers. A brand targeting the EU, US, and China simultaneously will hit a lower practical ceiling on alpha-arbutin than a brand targeting EU only — not because the science changes, but because the NMPA approval pathway for higher concentrations in special-use categories adds timeline and cost that most launch schedules can’t absorb.
One thing to be clear about up front: if China whitening registration is in scope, we need to know that at brief stage, not after stability is underway. The evidence package for NMPA special-use registration differs enough from a standard cosmetic dossier that it changes our testing plan from week one. Specifically, a human efficacy trial conducted at a Chinese CRO should be initiated within the first 8 weeks of the project if you’re targeting a 12-month China launch.
On timeline: lab samples in 2 to 3 weeks, accelerated stability over 4 to 8 weeks, 24-month real-time stability initiated concurrently. If ISO 24444 SPF testing is required, add 8 to 12 weeks and plan that against formula lock, not stability completion.
Frequently Asked Questions #
We want to sell in both Europe and China under the same SKU — is that actually achievable?
A: It depends entirely on your claim language. A cosmetic claim that avoids any direct reference to melanin or skin color lightening can often clear both markets — EU as a standard cosmetic notification, China as a general cosmetic (non-special-use) with appropriate framing. The moment you want to make an explicit whitening claim for the China market, you’re looking at special-use registration, which requires a China-specific evidence package and 9 to 15 months pre-market. Same SKU, different claims on different market labels, is the approach most brands end up taking.
Is kojic acid still viable for EU formulation after the SCCS review?
A: Yes, but at reduced limits. The current position following the SCCS Scientific Opinion restricts kojic acid to 1% in face rinse-off products and 0.7% in face leave-on products. Body leave-on application is still under review — we don’t recommend building a body brightening formula around kojic acid right now until that opinion is finalized, because a product compliant today could require reformulation if the limit drops. For body applications, we’d steer you toward alpha-arbutin or tranexamic acid for EU market stability.
What’s the stability watch-out with brightening actives at the concentrations required for EU compliance?
A: The real issue is oxidative degradation, not concentration. We’ve had batches of 2% alpha-arbutin serum show acceptable accelerated stability at 40°C for 8 weeks, then fail at week 12 in real-time due to chelation failure in the water phase. The culprit in those cases was trace metal contamination from manufacturing equipment, which we now screen for under our incoming water quality protocol. For vitamin C derivatives, discoloration at anything above pH 4.0 is a consistent challenge — it doesn’t affect safety but it fails consumer acceptance testing, which is just as disqualifying for EU retail.
What’s a realistic MOQ and timeline for a multi-market brightening product?
A: For a standard serum or cream format, MOQ in our facility is typically 500 kg per SKU. Timeline from signed brief to first lab samples is 2 to 3 weeks; accelerated stability completion at 4 to 8 weeks; finished regulatory dossier (EU CPNP-ready) at 12 to 16 weeks from formula lock. If China NMPA special-use registration is in scope, that runs in parallel but adds 9 to 15 months before China market launch. EU and US can typically launch in 5 to 6 months from brief — China is the constraint on any multi-market schedule.
Should we align our formula to ISO 16128 natural index even if we’re not making a “natural” claim?
A: This is a question more brands should ask upfront, and the answer depends on your channel strategy. If you’re targeting EU mass retail or clean beauty platforms, many buyers now request ISO 16128 documentation even without a formal on-pack claim — it’s part of their own brand sustainability reporting. But for a high-performance brightening and whitening formula targeting clinical-efficacy positioning, aligning to ISO 16128 often forces you away from the most effective synthetic actives. We’re still not convinced the tradeoff is worth it for every brand. Our recommendation: know what your retail buyer requires before you scope the formula, not after you’ve already selected actives.
Have a product concept in mind? Contact our formulation team to request a complimentary brief review.
The China NMPA dossier timing is what catches teams off guard most often — we’ve had formulas where the stability protocol was solid for EU submission, but the human trial data requirements for NMPA registration added 14 months to the timeline because we didn’t scope that in at brief stage.
MOQ reality that doesn’t get talked about enough: most OEMs with NMPA-registered whitening actives on their approved list require 5,000-unit minimums per SKU, which sounds manageable until you’re running three separate regional variants of the same product because your niacinamide concentration had to drop for the EU label and your claim language had to be stripped entirely for the US version.
The quasi-drug classification in Japan is one that still trips up EU-native brands — niacinamide as a brightening active sits fine in an EU cosmetic at 5%, but the moment you’re filing a Quasi-Drug Application with PMDA, you’re working from a fixed approved ingredient list where the permitted concentration and intended function are locked by the Ministry. We had a formula using 4% niacinamide plus alpha-arbutin that sailed through CPNP notification but needed a full reframe for Japan because the combination claim read as quasi-drug territory to the local RA consultant.
We briefed our OEM in Guangzhou on a dual-market formula targeting EU and China, and they built the entire stability protocol around our EU CPNP submission timeline — 12 months accelerated, standard conditions. Nobody flagged that the NMPA registration for the whitening claim would need a separate efficacy dossier with a clinical brightness assessment that took another 8 months on top. We launched EU on schedule and sat on 18,000 units of finished goods in a bonded warehouse for the better part of a year waiting for China clearance. The cost of that delay, just in storage and capital tied up in stock, was more than the original formula development budget.
Phenylethyl resorcinol at 0.5% was our chosen brightening active for a three-market launch, and the delta between our EU CPNP notification (6 weeks) versus NMPA special-use cosmetic registration (11 months) meant we ended up with a split launch strategy we hadn’t budgeted for.
Claim language being the real driver of classification burden tracks exactly with what we ran into on a vitamin C + alpha-arbutin serum last year — kept the arbutin at 1.8% specifically to stay under the EU 2% face limit, but the moment our marketing team drafted “reduces dark spots by 40%” for the US copy, the product started looking a lot more like a drug claim than a cosmetic one and we had to walk that entire brief back to “brightens the appearance of uneven skin tone” before anyone downstream would touch it.