Vegan cleaning products, candles and glue: the aisle cosmetics rules don't reach

In 2013 the European Union banned selling any cosmetic whose finished product or ingredients had been tested on animals anywhere in the world, under Article 18 of the Cosmetics Regulation, EC 1223/2009. That ban reshaped an entire industry and is the reason cruelty-free became a normal shelf claim in beauty aisles. It has never applied to the bottle of dish soap next to the sink. Cleaning products fall under a separate law, the Detergents Regulation, EC 648/2004, which historically said nothing about animal testing at all. A revision changes that, but only from 23 September 2029, when a ban on animal testing for detergents and their ingredients finally takes effect across the EU, adopted by the European Parliament on 11 February 2026. For the thirteen years between the cosmetics ban and that date, a shampoo and a shower gel could carry a cruelty-free promise the all purpose cleaner on the same shelf legally could not.

The gap does not close cleanly even in 2029. Both regulations sit on top of REACH, the EU's general chemical safety law, and REACH treats animal testing as a last resort rather than a ban. A company registering a surfactant or fragrance chemical above certain tonnage thresholds must show it considered non-animal methods first, but regulators can still require a vertebrate test to establish how a substance behaves in a factory worker's lungs or in a river, uses that have nothing to do with whether the finished bottle was tested on a rabbit's skin. So the same ingredient can be legally testing free as a cosmetic component and still be subject to an animal test as a registered industrial chemical. A brand can accurately claim its finished product was never tested on animals while an ingredient supplier further up the chain was required to test it for an unrelated safety file.

Bar soap is where this shows up most literally. Sodium tallowate, the sodium salt of rendered beef or mutton fat, is made by boiling tallow with lye in the same saponification reaction that has made soap for centuries, and it produces a harder, longer-lasting bar with a richer lather than most plant oils manage alone. It shows up on the label by that exact name, so it is one of the easier animal ingredients in this whole category to actually catch. Vegetable oil soaps use sodium palmate, sodium cocoate or sodium olivate instead, made the same way from palm, coconut or olive oil.

Glycerin is the harder case, because the molecule is chemically identical whether it started in an animal or a plant. It falls out of the same saponification reaction as a byproduct, whether the fat going in was tallow or soybean oil, and once refined there is no test that tells you which. Manufacturers know the source because they buy the input, but nothing on a bottle of hand soap requires them to say. The same is true of stearic acid, a saturated fatty acid that makes up roughly a fifth of beef tallow by weight and is also present in coconut and palm oil. Magnesium stearate, its calcium and magnesium salts, and most fatty acid derivatives with names ending in -ate or -amide carry the same ambiguity. The full verdict page covers how to press a manufacturer for a straight answer.

Lanolin, the grease sheep secrete into their wool, is less hidden but often unrecognized. It is the ingredient that makes leather conditioners and some furniture and shoe polishes feel almost like skin cream, because its molecular structure closely resembles the natural oils in animal skin and it penetrates leather fiber rather than sitting on the surface the way a synthetic polymer does. That is also exactly why it works so well and why plant oil substitutes, usually a blend of carnauba wax and vegetable oils, need more frequent reapplication to match it.

The proteolytic enzymes that break down protein stains in modern laundry detergent are not animal derived today, and that is a real change from where the industry started. The first commercial enzyme detergent, launched in Germany in 1913, used trypsin extracted from ground animal pancreases, a crude and inconsistent process. Almost every enzyme in laundry detergent since the 1960s has instead been grown from bacteria, chiefly Bacillus subtilis and Bacillus licheniformis, through industrial fermentation, and proteases now account for roughly 60 to 65% of the global industrial enzyme market, most of which is sold into detergents.

The genuinely unresolved question is what feeds the bacteria. A typical fermentation broth for producing these enzymes uses glucose and soybean meal, both plant derived, but the growth medium is proprietary to each manufacturer and can vary by batch and by plant, and the industry does not publish it on the finished detergent. The enzyme protein itself is vegan by any reasonable definition once purified. Whether the microorganism that produced it was ever fed anything animal derived along the way is not something an ingredient list can answer, and it is a smaller, more theoretical version of the same disclosure gap that runs through this whole category.

A traditional candle is two animal materials working together. Beeswax alone burns cleanly and holds scent poorly, tallow alone burns dirtier and smells of rendered fat, so for most of candle history the fix was stearic acid, extracted from tallow, added to harden the wax and raise its melting point. That combination is why old tallow candles smoked and modern wax blends do not. Pure beeswax candles are still sold on the strength of a naturally sweet scent, but beeswax does not hold added fragrance oils well and needs a much higher fragrance load to throw scent the way a paraffin or coconut wax candle does at a fraction of the concentration.

The plant alternatives split by what they are good at. Soy wax is the cheapest and most common, with a good but not exceptional scent throw. Coconut wax throws scent nearly as strongly as paraffin while burning cleaner, but it costs two to three times as much per pound and is often blended with soy to control cost. Rapeseed wax gives a stable, low-soot burn with a creamy texture close to beeswax and is popular in the UK and EU market specifically because it is a domestic crop. Palm wax makes striking crystalline finishes but carries the same deforestation and orangutan habitat concerns as palm oil in food, so a candle can be entirely animal free and still sit on the wrong side of an environmental argument, which is the honest complication running through most of this category: the vegan choice and the sustainable choice are not automatically the same purchase.

Animal hide glue, boiled down from collagen in skin, bones and connective tissue, was the dominant wood adhesive for most of industrial history and is still specified by furniture restorers and some violin makers because it can be reversed with heat and moisture, a property synthetic glues do not share. Casein glue, made from milk protein and lime, was the standard structural wood adhesive through the early twentieth century, strong enough that it bonded the plywood used in early aircraft. Both have been almost entirely replaced in general household and craft use by polyvinyl acetate, the synthetic polymer in ordinary white and wood glue, along with epoxies and cyanoacrylates, none of which involve animal material. If a glue bottle just says "wood glue" or "all purpose" without specifying hide or casein, it is almost certainly synthetic.

Paint tells a similar story in reverse. Casein paint, sometimes marketed today as milk paint, uses the same milk protein binder that made casein glue useful, and it is having a small revival among natural-finish decorators precisely because it produces the matte, breathable look of traditional distemper without the animal glue that older distemper recipes used as a binder. Shellac still does real work in wood finishing and floor sealing, sold as flakes or in solution, and it is entirely insect derived: the female lac beetle secretes the resin as a protective cocoon around her eggs, and confectioner's glaze, the food-grade version used to give candy and pill coatings their shine, is about 35% of that same resin. The shellac verdict page covers the food-label side of this in detail.

Three certification schemes reach into this category, and most shoppers know only the one that appears on cosmetics. Leaping Bunny, run in North America by the Coalition for Consumer Information on Cosmetics since 1996, certifies household cleaning products alongside personal care and pet products under its Corporate Standard of Compassion for Animals, requiring a company to prove its entire supply chain, including raw material suppliers, is free of animal testing, and to submit to an annual re-audit to keep the certification current. PETA's list, first launched in 1987 as Beauty Without Bunnies and since renamed, covers household cleaning products explicitly and carries two separate marks: "Animal Test-Free" certifies no testing anywhere in the supply chain, while "Animal Test-Free and Vegan" additionally requires the entire product line to contain no animal-derived ingredients. Listing is free; a company pays a one-time $350 fee only if it wants to print PETA's logo on the package.

The Vegan Society's Trademark, in use since 1990 and licensing more than 70,000 products worldwide, is the strictest of the three on ingredients because it explicitly extends to processing aids, substances used during manufacturing that never appear on the final ingredient list, such as a filtering agent. None of the three schemes overlaps completely with the others, and none of them is required by law to appear on a household product the way, say, a hazard pictogram is. The beauty and cruelty-free labels guide covers how these same three marks behave on cosmetics, where the underlying law is different and, since 2013, considerably stricter.

"Not tested on animals" printed directly on a bottle, with no certification logo attached, is a self-declared claim with no independent audit behind it in most jurisdictions. It usually means the finished product was not tested, which says nothing about the individual ingredients or their suppliers, and under REACH it can be true of the bottle and false of a chemical inside it at the same time.

Most mainstream liquid detergents, glass cleaners and all purpose sprays sold today are already free of animal-derived ingredients even without a vegan logo, because the petrochemical and plant-oil surfactants that dominate modern formulations are cheaper and more consistent to source than tallow derivatives. The exceptions cluster in specific product types: bar soap, anything marketed as "natural" or "traditional" which more often means tallow rather than less, candles, leather and furniture care, and craft glue and paint. That narrows the label reading to a short list of terms worth scanning for: tallow, sodium tallowate, glycerin without a "vegetable" qualifier, stearic acid or magnesium stearate without a plant source stated, lanolin, beeswax, casein, gelatin and shellac.

The most useful shortcut in this entire category costs nothing. White vinegar, bicarbonate of soda and citric acid, all mineral or plant derived and none of them ever formulated with an animal ingredient, replace glass cleaner, scale remover, fabric softener and a general degreaser between them, and none require a certification because there is no ambiguous ingredient to certify.

Certification coverage in this category is thin compared to food or even cosmetics. A large share of household brands, especially store brands and industrial-use products, carry no certification at all in either direction, not because they contain animal ingredients but because the audit costs money and household cleaning has never faced the consumer pressure that made cruelty-free a marketing necessity in beauty. That means the honest answer to "is this cruelty free" for most of what is under your sink is that nobody has checked, not that the answer is no. A reader who wants certainty on a specific product, rather than a general steer, has to write to the manufacturer and ask directly about tallow-derived ingredients and their glycerin and stearic acid sourcing, because the label was never going to volunteer it.