What Glycerin Actually Does

Glycerin sits near the top of an enormous number of ingredient lists because it is a humectant — it holds water in the outer layer of skin. In 2014 it was the third most frequently reported cosmetic ingredient in the FDA's voluntary registration database, after water and fragrance.

That part is uncontroversial. The interesting part is a claim you have almost certainly read: that in dry air, glycerin reverses and pulls water out of your skin. We went looking for the study behind it.

There isn't one.

Where the low-humidity claim actually comes from

The sentence circulates across the beauty internet in near-identical form. Its earliest traceable appearance is not in a study. It is in the background section of a patent — the part where an applicant describes the problem their invention supposedly solves:

"when humectants are used in low humidity air, such as in late winter or early spring, they will often draw moisture away from the skin itself thus causing undesirable dryness."

That passage carries no experimental data and cites no literature. A patent background is an assertion written to motivate the applicant's own invention. It is close to the weakest provenance a factual claim can have.

The human evidence points the other way

If glycerin were siphoning water outward, you would expect transepidermal water loss — the standard measure of how much water is escaping through the skin — to go up.

Fluhr and colleagues (1999) tested glycerol on human skin using two separate damage models: tape-stripping, and repeated washing with sodium lauryl sulfate. In both, glycerol-treated sites showed faster barrier repair measured by transepidermal water loss — and the difference against untreated sites was still present seven days after treatment stopped. The authors concluded glycerol "can be regarded as a barrier stabilizing and moisturizing compound."

That persistence is worth noticing. A week after you stop applying something, simple surface occlusion has long since washed off. It argues for a structural effect rather than a film.

Humidity does matter — in the opposite direction

Here is the part that makes the folklore especially frustrating: dry air genuinely does change what glycerin does. It just does not do what the story says.

Rawlings and colleagues (1995) found glycerol accelerates desquamation — the shedding of dead surface cells — and that this effect is humidity-dependent, "being significantly reduced at low relative humidity."

So low humidity blunts a benefit glycerin provides. It does not flip glycerin into a water thief. The real humidity story is the inverse of the popular one.

What glycerin does besides hold water

This is the genuinely under-reported half of the story.

It helps skin shed properly. Rawlings found desmosomes — the protein rivets holding surface skin cells together — at more advanced stages of degradation in glycerol-treated tissue than in control, corroborated by three independent measures. In scaly dry skin the actual defect is that dead cells fail to detach on schedule. Glycerin is not only holding water there; it is helping the shedding happen.

It repairs barrier function. The Fluhr result above, in humans, across two insult models.

It changes the lipid matrix. A 2024 paper in Biophysical Journal working with isolated human stratum corneum lipids found glycerol increased the cohesion of the lipid layers and stabilised their ordered domains — without altering the overall lipid structure or how permeable it was.

The mechanism is less settled than you would think

Two caveats that most ingredient explainers skip.

The famous transport mechanism is mouse data. Glycerin's movement in skin is usually attributed to aquaporin-3, a channel that carries glycerol. The causal work establishing that comes from knockout mice, where deleting AQP3 depleted glycerol specifically in the outer skin layers and not in dermis or blood. A companion study showed the resulting hydration deficit corrected after glycerol was given — but it corrected whether the glycerol was applied topically or given by mouth or injection. That makes it good evidence that glycerol matters physiologically, and weaker evidence about what a cosmetic does when you rub it on. We could not obtain a human study establishing this pathway.

And one 2024 result complicates "humectant" itself. The Biophysical Journal work found that in isolated human stratum corneum lipids, glycerol did not increase water sorption, and the authors described it as increasing cohesion "rather than acting as a humectant." That does not overturn the category — they deliberately excluded the skin cells themselves, where classic humectant action is thought to occur, and they say so. But it means the mechanism is less nailed down than the confident one-liners suggest.

There is no "typical" concentration

If you have seen an article state a typical glycerin percentage, it was invented. The Cosmetic Ingredient Review's safety assessment reports an industry survey of maximum use concentrations spanning 0.0001% to 79.2% in leave-on products — nearly six orders of magnitude. Glycerin turns up as a trace solvent and as very nearly the whole product.

In the US it crosses the cosmetic/drug line

A detail almost no consumer coverage mentions. In the United States, glycerin is not only a cosmetic ingredient — at particular concentrations it is a regulated over-the-counter drug active ingredient.

21 CFR 347.10, the list of skin protectant actives, includes:

"(h) Glycerin, 20 to 45 percent."

It sits in that list beside allantoin, calamine, cocoa butter, dimethicone and kaolin. Below 20% or above 45%, glycerin is not a monograph skin protectant — it is back to being a cosmetic ingredient. The same molecule changes regulatory category with its concentration.

The CIR panel's conclusion on the cosmetic use is that "glycerin is safe in cosmetics in the present practices of use and concentration described in this safety assessment."

What we could not establish

• The EU database entry. CosIng, the EU cosmetic ingredient database, is a JavaScript application that serves an empty shell to anything but a real browser. Three URL forms returned no data. We therefore have no CosIng-sourced fact in this article, and we have not published an EC number — one appeared in a search snippet and was never seen on a page we fetched.
• Whether the AQP3 pathway operates the same way in humans. The causal evidence is mouse knockout work. One relevant human study was behind a bot check on repeated attempts.
• A direct test of the low-humidity claim. As above: unsupported and contradicted in direction, not disproved by experiment.
• Which of the CIR report's two sets of use-level figures is correct. They are in the same document and disagree.

Sources consulted, retrieved 28 July 2026:

• Cosmetic Ingredient Review, Safety Assessment of Glycerin (Final Report, January 2015). PDF text extracted locally and each quotation confirmed present. Cited for cosmetic function, CAS number, use-concentration survey and the panel's conclusion.
• 21 CFR 347.10 — skin protectant active ingredients, read via the eCFR versioner API. Quoted directly rather than via secondary reproduction.
• Rawlings A, Harding C, Watkinson A, et al. The effect of glycerol and humidity on desmosome degradation in stratum corneum. Arch Dermatol Res 1995.
• Fluhr JW, et al. Glycerol accelerates recovery of barrier function in vivo. Acta Derm Venereol 1999;79(6):418–21.
• Fluhr JW, Darlenski R, Surber C. Glycerol and the skin: holistic approach to its origin and functions. Br J Dermatol 2008;159(1):23–34.
• Hara M, Ma T, Verkman AS. J Biol Chem 2002, and Hara M, Verkman AS. PNAS 2003 — the aquaporin-3 knockout work, in mice.
• Sagrafena I, et al. Biophys J 2024 — glycerol and isolated human stratum corneum lipids.
• US Patent 10307366B2, "Description of the Related Art". Cited only as the traceable origin of the low-humidity claim, not as evidence for anything. A patent background is an assertion made to motivate its own invention.
• Zeichner JA, Del Rosso JQ. J Clin Aesthet Dermatol 2016 — cited as an example of the claim being repeated without a supporting reference, not as support for it.

Presence of an ingredient does not establish that a product works: concentration, the rest of the formula, pH and delivery all matter, and none of them are on the label. This is general information, not medical advice. GlowSteal has not tested any product for this article.