Is Phenoxyethanol Bad for Your Skin? The 1% Rule — and How to Read the Dose Off the Label

By S. C. Aris — about the author

Last updated: 16 August 2026

What phenoxyethanol actually is — and why it’s in your bottle

Phenoxyethanol (CAS 122-99-6, sometimes written 2-phenoxyethanol or ethylene glycol monophenyl ether) is an oily, faintly rose-scented liquid used as a broad-spectrum preservative. It has been in cosmetics since the 1950s and is now one of the two most common preservatives on the shelf. It appears in moisturisers, serums, sunscreens, foundations, cleansers, baby wipes and micellar waters.

It is there for an unglamorous reason: anything containing water will eventually grow something. The US Food and Drug Administration’s guidance on microbiological safety in cosmetics lists the routes in: contaminated raw materials or water, poor manufacturing conditions, an “ineffective preservation system particularly for ingredients that encourage growth of microorganisms”, packaging that doesn’t protect the product, bad storage — and, pointedly, “consumer use, such as dipping fingers into the product.”

The FDA also notes that while injuries from contaminated cosmetics are not common, “they can be serious”, citing contaminated tattoo inks, eye-area products and hospital lotions and mouthwashes as causes of real infections. Its consumer advice includes not sharing products, not adding water or saliva to them (which “waters down preservatives”), storing them as instructed, and washing your hands before dipping fingers in.

So the question is never “preservative or no preservative.” It is “which preservative, at what dose, in what kind of formula” — or, alternatively, “a formula with no water in it at all.”

What the regulators actually concluded (the full record, in one table)

Most articles on this topic quote one line from one body. Here is the complete assessment history, with what each one actually looked at.

One caveat the SCCS itself flags, and which is worth knowing: its opinion explicitly states that it “does not take into account exposure from sources other than cosmetics.” That is an honest limitation, not a hidden danger — but it is the sort of thing a genuinely useful article should tell you rather than bury.

Sources: SCCS Opinion on Phenoxyethanol, SCCS/1575/16 (European Commission); the Commission-published commentary carrying the final conclusion verbatim; and the CIR Expert Panel conclusions as summarised on Cosmetics Info — a database run by the Personal Care Products Council, i.e. the industry, which we flag so you can weight it accordingly. The CIR reports themselves are linked from that page.

Where the fear actually came from

Three separate things get blended into one internet claim. Pulling them apart is most of the answer.

1. The 2008 nipple-cream warning — which was about swallowing, not skin

In 2008 the FDA warned consumers not to use one specific product, Mommy’s Bliss Nipple Cream, because two of its ingredients — chlorphenesin and phenoxyethanol — could reportedly cause respiratory distress, or vomiting and diarrhoea, in infants exposed to it while nursing. The product was voluntarily withdrawn. The FDA had not received reports of illness linked to it.

Read the exposure route carefully: this was a cream applied to a nipple, from which a newborn then fed. That is oral dosing of a very small person, repeatedly, with two central-nervous-system-active ingredients together. It tells you nothing about whether a 0.7% phenoxyethanol moisturiser is safe on an adult’s cheek. Almost every “the FDA warned about phenoxyethanol” headline you will see omits the word “nipple.”

2. The French proposal — which was about nappies

ANSM’s 2012 risk assessment, quoted in the SCCS opinion’s own background section, concluded that the maximum should be lowered to 0.4% for children under three and that phenoxyethanol should not be used in products intended for their nappy area. The nappy area is occluded, often broken skin, on the youngest and smallest users — a worst-case exposure scenario by design. The Commission took the concern seriously enough to commission a full re-review, which is exactly how the system is supposed to work. That re-review is SCCS/1575/16, above.

3. The animal data — which was run at doses nobody encounters

The SCCS dossier contains the studies people quote out of context. In a 13-week dermal study in rabbits — the species the SCCS calls “the most sensitive species to haematotoxic effects when systemically exposed to 2-phenoxyethanol” — undiluted phenoxyethanol was applied under an occlusive bandage for six hours a day, five days a week, at up to 500 mg/kg of body weight per day. At that dose the no-observed-adverse-effect level was 500 mg/kg bw/day; the SCCS adjusted it to 357 mg/kg bw/day for the dosing schedule. For an adult, that is roughly the equivalent of strapping tens of grams of neat phenoxyethanol to your skin under a dressing every weekday. A 50 ml moisturiser at the legal maximum contains half a gram in the whole bottle.

This is the pattern behind almost every “linked to organ damage” claim about a cosmetic preservative: a real study, a real finding, and a dose several hundred-fold away from the product on your shelf.

The label position test: reading the dose off the bottle

Here is the piece the other results are missing. You cannot see a percentage on a cosmetic label — but US labelling law does let you bracket it, if you know where the boundary sits.

Under 21 CFR 701.3, the ingredient declaration must list each ingredient “in descending order of predominance”. But paragraph (f) gives manufacturers an alternative they almost all use:

“(1) Ingredients, other than color additives, present at a concentration greater than 1 percent, in descending order of predominance; followed by (2) Ingredients, other than color additives, present at a concentration of not more than 1 percent, without respect to order of predominance; followed by (3) Color additives, without respect to order of predominance.”

That single clause is the whole trick. It means an ingredient list has an invisible line through it. Above the line, order means concentration. Below the line, order means nothing at all — a 0.9% ingredient can legally be listed after a 0.05% one.

How to run the test

1. Find the anchor ingredients. Look for the ones that are essentially always used at or below 1%: phenoxyethanol itself, fragrance/parfum, tocopherol, disodium EDTA, xanthan gum, ethylhexylglycerin, sodium hydroxide, citric acid, and the named fragrance allergens (limonene, linalool, geraniol, citronellol).
2. Draw the line at the first anchor. Everything from the first reliable sub-1% anchor onward is the “1% and under” block. In practice, the line often sits right around where the first preservative or thickener appears.
3. Read phenoxyethanol’s side of the line. If it sits below the line — which it will in essentially every compliant product, because 1% is the legal ceiling anyway — you know it is at 1% or less, and you know its exact position within that block tells you nothing. Someone claiming a product is “loaded” with phenoxyethanol because it’s listed above tocopherol is reading noise.
4. Check whether it is above the line. If phenoxyethanol appears before obvious high-percentage ingredients like glycerin or a butter, that is worth a second look — it would imply an unusually high level, and in an EU-market product it would be non-compliant above 1.0%.
5. Count the co-preservatives. Phenoxyethanol is often paired with ethylhexylglycerin, caprylyl glycol, chlorphenesin, sodium benzoate or potassium sorbate. Seeing two or three of these together is normal formulation, not a red flag — it usually means each is used at a lower level than one would be alone.

Two honest limits on the test. First, 21 CFR 701.3(l) exempts “incidental ingredients” — substances present at insignificant levels with no functional effect, including those carried in as part of another ingredient — from declaration entirely. Second, this is a US rule; the EU has its own labelling framework under Article 19(1) of Regulation (EC) 1223/2009, under which perfume compositions are declared simply as “parfum” or “aroma” with only a defined list of allergens named individually. So the test brackets the dose; it does not measure it.

Preservative systems compared — including the option of not having one

Most comparisons stop at “phenoxyethanol vs parabens.” There is a third column that changes the question.

That last row is where the “should I avoid preservatives?” question actually resolves. There is no virtue in a water-based cream with no preservative — that is a contaminated product waiting to happen, and the FDA’s contamination guidance above explains why. But a formula that never contained water never needed one.

“Phenoxyethanol-free”: what usually replaced it

When a brand removes phenoxyethanol from a water-based product, it has to put something else in. In practice you will typically find one of these on the new label:

• Sodium benzoate + potassium sorbate — the common “natural-positioning” swap. Requires the formula to sit at a low pH to work properly.
• Benzyl alcohol, often with dehydroacetic acid — effective, and itself a substance on the EU’s list of individually labelled fragrance allergens.
• Caprylyl glycol, 1,2-hexanediol, glyceryl caprylate — multifunctional humectant-solvents with mild antimicrobial activity, frequently marketed as “preservative-free” because they are not classified as preservatives even though they are doing preservative work.
• Chlorphenesin — worth noting, because it was the other ingredient in the 2008 nipple-cream warning, and the one the FDA described as able to depress the central nervous system in infants.

None of that makes any of these bad. It does mean “phenoxyethanol-free” on the front of a water-based bottle is a marketing statement about one molecule, not a statement about preservation. The only labels where “preservative-free” is structurally rather than semantically true are the water-free ones.

Who genuinely should be cautious — and how to test it yourself

Phenoxyethanol is a rare sensitiser, not a common one. But rare is not never, and if you are one of the people who reacts, the population statistics are cold comfort. Consider ruling it out if:

• You have persistent facial or hand dermatitis that flares with leave-on products and settles when you stop everything.
• You have atopic dermatitis or a compromised barrier — broken skin changes the exposure picture for any ingredient.
• You already patch-test positive to other preservatives or to Fragrance Mix I.
• You are choosing products for a baby’s nappy area, where ANSM’s specific concern was directed.

The 14-day repeated open application test (ROAT)

Formal diagnosis is a dermatologist’s patch test. But the repeated open application test, as described by DermNet, is the recognised at-home way to work out whether a specific leave-on product is the culprit:

1. Choose a small patch of skin on the inner forearm or inner upper arm.
2. Apply a dot of the suspect leave-on product to the same spot twice daily for two weeks.
3. Watch for redness, itching, scaling or small bumps developing over days — allergic contact dermatitis is a delayed reaction, appearing hours to days after exposure, not within minutes.
4. If nothing appears in fourteen days, that product is very unlikely to be your trigger.

DermNet notes the test is not suitable for undiluted wash-off products. It also makes two points that apply to every “free-from” label you will ever read: products marked “fragrance-free” or “unscented” “could still include masking fragrances”, and ingredients are only required to be listed above a specified threshold, so “manufacturers may not list all”. Elimination by label reading alone has limits. Elimination by testing does not.

The water question: why some serums don’t carry a preservative at all

Look at the first ingredient on most serums and you will see Aqua. That is not a criticism — water is a legitimate and often necessary solvent. But it is the reason the preservative conversation exists at all, and it has a second consequence for how the product behaves on skin.

Water evaporates. The lipids in a formula do not. In their 2018 review of topical plant oils in the International Journal of Molecular Sciences, Lin, Zhong and Santiago describe the stratum corneum as a “brick wall” of corneocytes held in intercellular lipid lamellae of ceramides, free fatty acids and cholesterol. They note that many plant lipids largely remain at the skin surface, that monounsaturated free fatty acids such as oleic acid act as permeability enhancers, and that the ratio of free fatty acids to triglycerides correlates with how far a lipid travels. They also record that linoleic-rich sunflower seed lipid preserved stratum corneum integrity and improved hydration, and that the occlusive effect of plant oils reduces transepidermal water loss even without penetration.

The practical upshot: a water-free lipid concentrate is a different delivery format, not just a preservative-free one. It has no aqueous phase to preserve, and nothing in it evaporates off the skin thirty seconds after application. We covered the evaporation half of this in what penetration studies actually show about face oils.

Frequently asked questions

Is phenoxyethanol safe for babies and children?

The SCCS addressed this directly. Having reduced its toxicokinetics default factor, it calculated a margin of safety of about 50 for children and stated that this “also covers this specific age group who might be higher exposed to 2-phenoxyethanol than adults” — at the 1.0% maximum. The separate, narrower concern raised by France’s ANSM in 2012 was specifically about products for the nappy area of children under three, and about lowering the cap to 0.4% for that age group.

Is phenoxyethanol safe during pregnancy?

The SCCS reviewed reproductive and developmental toxicity data as part of its 2016 assessment and concluded the ingredient is safe at up to 1.0% without carving out an exception for pregnancy. That said, its opinion explicitly excludes exposure from non-cosmetic sources, and pregnancy skincare questions are worth raising with your midwife or doctor rather than settling from an article.

Does phenoxyethanol cause cancer?

No regulatory body assessing it for cosmetic use has classified it as carcinogenic. The CIR Expert Panel found it was not genotoxic and raised no systemic-toxicity concern; the SCCS reviewed the full toxicological dossier and concluded it is safe at 1.0%. Claims to the contrary generally rest on high-dose animal studies of the kind described earlier in this article.

Can phenoxyethanol cause an allergic reaction?

Yes, but rarely. The CIR Panel did not classify it as a skin sensitiser, and case reports of allergic contact dermatitis and contact urticaria to it are uncommon in the literature relative to how widely it is used. If you suspect a reaction, the 14-day ROAT above is the practical first step, and a dermatologist’s patch test is the definitive one.

How can I tell how much phenoxyethanol is in a product?

You cannot get an exact figure, but you can bracket it. Under 21 CFR 701.3(f), ingredients above 1% must be listed in descending order and ingredients at 1% or less may be listed in any order. Find the first reliable sub-1% ingredient on the list; everything after it is in the 1%-and-under block, and position within that block carries no information about concentration. In the EU the legal ceiling is 1.0% regardless.

Is a preservative-free product better?

Only if it never needed one. A water-based product without adequate preservation is a contamination risk, which is exactly what the FDA’s microbiological safety guidance is about. A water-free formula — a lipid serum, a balm, an anhydrous concentrate — has no aqueous phase for microbes to grow in, so “preservative-free” is a structural fact about it rather than a claim.

Should I throw out my products that contain phenoxyethanol?

On the evidence, there is no reason to. If you prefer to avoid it, the honest route is to move to water-free formats rather than to water-based products that have simply swapped in a different preservative under a less familiar name.

The bottom line

Phenoxyethanol is one of the most thoroughly assessed ingredients in your bathroom, and both the US and EU expert bodies that have looked at it have cleared it for cosmetic use. The alarming-sounding evidence against it is real evidence about a different exposure: infants swallowing it, occluded nappy-area skin, or animal studies at doses hundreds of times higher than a cosmetic delivers. A small number of people do develop a genuine contact allergy, and for them a 14-day ROAT is worth more than any ingredient list.

If you would rather not have a preservative in your routine at all, the answer is not to look for a water-based product that has dropped one molecule. It is to look at whether the product contains water in the first place.

This article is for general information about cosmetic products and their ingredients. Our products are cosmetics and support the appearance and feel of skin; they are not intended to diagnose, treat, cure or prevent any medical condition. If you have persistent dermatitis, a suspected allergy, or any skin concern that is changing or not settling, please see a doctor or dermatologist.