What's Actually in Your Skincare and Medicine Cabinet

TL;DR. Rub in a night cream and reach for an acne wash the next morning, and you've crossed between two entirely different regulatory worlds without noticing. In the United States, a cosmetic like that night cream can legally reach a store shelf with no safety data filed with any regulator at all; the company selling it is simply required to believe, privately, that it's safe. The acne wash, the dandruff shampoo, and the sunscreen sitting next to it are something else entirely: over-the-counter (OTC) drugs, reviewed under a decades-old rulebook called the monograph system, a pathway between plain cosmetics and the multi-phase prescription drug pipeline covered in an earlier chapter. Cosmetics went almost entirely unregulated at the federal level from 1938 until 2022, a gap patched only by an industry-funded safety panel and a patchwork of state laws, until a new law finally gave the FDA power to order a recall instead of just asking a company nicely.

Key takeaways

  • Cosmetics, OTC drugs, and prescription drugs are three legally distinct FDA categories under the same 1938 statute, with three very different requirement sets. Cosmetics have historically needed no premarket approval at all; OTC drugs go through the "monograph" system; prescription drugs go through the trial-and-approval pipeline this book already covered.
  • Sunscreen is legally an OTC drug in the U.S., not a cosmetic, and is tested for SPF and broad-spectrum protection under FDA-mandated methods, even though almost everyone shops for it next to the body lotion.
  • A cream or lotion is chemistry as much as marketing: an emulsion of oil and water phases held together by an emulsifier, carrying active ingredients, preservatives, and stabilizers that each do one distinct job, applied to a skin barrier, the stratum corneum, that many actives are deliberately built to sit on top of rather than cross.
  • Modern cosmetics regulation exists because of the 1933 to 1938 "Lash Lure" eyelash dye tragedy, which blinded, disfigured, and in at least one case killed users, and helped push Congress into passing the Federal Food, Drug, and Cosmetic Act of 1938. Meaningful federal reform then didn't arrive again for another 84 years.
  • The 2022 Modernization of Cosmetics Regulation Act (MoCRA) gave the FDA mandatory recall authority, facility registration, product listing, and adverse event reporting for cosmetics for the first time, but critics point out it still doesn't require independent premarket safety testing or ban most of the ingredients the EU has restricted.
  • Marketing phrases like "dermatologist tested" and "clinically proven" have no standardized legal definition in the U.S. They can describe anything from a controlled clinical study to one dermatologist glancing at a formula sheet.

Two jars, two rulebooks

It's a routine two minutes: a dab of night cream worked into the cheeks and forehead before bed, and the next morning, a benzoyl peroxide acne wash, a dandruff shampoo, or a dollop of sunscreen rubbed into the neck before heading outside. Nothing about the ritual signals that these products were made and sold under completely different legal regimes. A bottle of face cream and a tube of acne treatment can sit on the same bathroom shelf, cost about the same, and be sold by the same drugstore chain, while one of them legally required no government review before it reached that shelf and the other went through a federally prescribed testing and labeling process before its manufacturer could call it "acne medicine" at all.

An earlier chapter in this book followed a prescription pill from a laboratory hypothesis through years of clinical trials to an FDA approval, a process that can take a decade and cost over a billion dollars per successful drug. Nothing here repeats that story; prescription drugs are their own category, and that chapter already covers it in depth. What sits right next to it, legally and physically, is a much less visible world: cosmetics and OTC personal care products, governed by rules that most people, including plenty who work in skincare marketing, have never had explained to them in plain language. The two products in your cabinet aren't just different formulas. They're different legal species.

What's actually happening in that jar

Peel back the marketing copy on a jar of face cream and what's left is applied chemistry with a long history. Most creams and lotions are an emulsion: a stable mixture of two liquids that don't naturally mix, an oil phase and a water phase, held together by an emulsifier, a molecule with one end attracted to oil and one to water, which coats tiny droplets of one liquid and keeps them suspended through the other instead of separating back out, the way oil and vinegar separate in a salad dressing left standing. In an oil-in-water emulsion, the more common type for lightweight lotions, oil droplets disperse through a continuous water phase, which is why the product feels light and rinses off easily. In a water-in-oil emulsion, more typical of heavier night creams and barrier ointments, the ratio flips, and the product feels richer and sits longer on the skin. A typical formula runs somewhere around 60 to 80 percent water, 10 to 25 percent oils and butters, and a few percent emulsifier, mixed in stages, an oil phase, a water phase, and the two combined under heat before the batch cools.

Layered into that base are the ingredients doing the product's actual work. Active ingredients are whatever is meant to change something, a retinoid meant to speed skin cell turnover, an alpha hydroxy acid meant to loosen the glue between dead surface cells, a mineral filter meant to scatter ultraviolet light. Preservatives exist for a less glamorous reason: the moment a jar or bottle is opened, it's exposed to air, bacteria on fingertips, and whatever else is floating around a bathroom, and a water-containing product without a working preservative system is, from a microbe's point of view, a warm, damp, nutrient-rich place to grow. And stabilizers, thickeners, and pH adjusters keep the emulsion from separating and keep the finished product's texture and appearance consistent from the day it's made until the day it's used up, often a year or more later.

None of that chemistry matters unless it can reach, or deliberately avoid reaching, its target, and skin makes both directions hard. The outermost layer of skin, the stratum corneum, is a thin layer of dead, flattened cells bound by a lipid-rich matrix, functioning less like a plain surface and more like a brick wall, dead cells as bricks, fats between them as mortar, extremely good at keeping water in and everything else out. Most molecules applied to skin never make it past that layer. For some products, that's exactly the point: a mineral sunscreen filter sits on top of the stratum corneum and scatters or absorbs UV light before it reaches living tissue, and a barrier ointment stays on the surface to seal moisture in. Other actives, particularly small, fat-soluble molecules like some retinoids, are formulated to work between those "bricks" and reach living cells below. Formulators decide, deliberately, which behavior they want; a large part of cosmetic chemistry is engineering a molecule's size and delivery system to either clear that barrier or respect it.

Cosmetic, OTC drug, or prescription drug: not the same thing at all

Here is the part that almost never gets explained on a store shelf. In U.S. law, a single statute, the Federal Food, Drug, and Cosmetic Act, defines cosmetics and drugs as two separate legal categories, and it's not the ingredient list or the price tag that decides which one a product is. It's the intended use, meaning what the label, packaging, and marketing claim the product does. A moisturizer that claims only to cleanse, beautify, or improve appearance is legally a cosmetic. A product that claims to treat, cure, mitigate, or prevent a disease, or to affect the structure or function of the body, is legally a drug, whether or not it needs a prescription to buy. That single distinction, appearance versus function, is why a jar of "anti-aging" moisturizer and a tube of "acne treatment" cream can sit inches apart on the same shelf while being regulated as if they came from different industries entirely, because in the eyes of the law, they did.

Cosmetics, historically, have needed no premarket approval whatsoever. Under the FD&C Act, a company can formulate a new lotion, manufacture it, and put it on shelves nationwide without submitting a single test result to the FDA first, with one narrow exception: color additives do require specific FDA approval before use. Everything else, the emulsifier, the fragrance, the preservative system, is the manufacturer's own responsibility to make sure is safe, and the FDA's authority has traditionally kicked in only after the fact, against a product already found adulterated or misbranded.

OTC drugs, nonprescription drugs sold directly to consumers, sit on a different track entirely, neither the cosmetics free pass nor the prescription drug's individualized, years-long review. Most OTC drugs, including acne treatments, dandruff shampoos, and sunscreens, are covered by the OTC monograph system, which the FDA began building in 1972. Rather than reviewing each company's product one at a time, the FDA sets a monograph, essentially a public rulebook for an entire drug category, listing which active ingredients are Generally Recognized as Safe and Effective (GRASE) at which concentrations, and what the product can claim and how it must be labeled. Any manufacturer can sell under a monograph without individual FDA review, as long as the formula and labeling match exactly. Step outside the monograph, with a new active ingredient or an uncovered claim, and the product needs its own full drug application instead.

Sunscreen is the clearest everyday proof this distinction is real, not academic. In the U.S., sunscreen is legally an OTC drug, not a cosmetic, because its label makes a functional claim: protecting skin from ultraviolet radiation. Every sunscreen sold in the U.S. has to be tested under FDA-mandated methods for SPF (a clinical, in-person test measuring how much longer treated skin takes to sunburn than untreated skin) and, since a 2011 rule finalized in 2018, for broad spectrum protection (a laboratory test confirming the product blocks both UVA and UVB radiation, not just the UVB rays that cause visible sunburn). A moisturizer on the same shelf that merely claims to "hydrate" needs none of that testing, which is exactly why a cosmetic moisturizer with SPF listed on the label is, legally, also classified and tested as an OTC drug the moment that SPF number appears.

Don't be confused: "cosmetic," "OTC drug," and "prescription drug" are not marketing categories, they're three separate legal boxes. A cosmetic is regulated for safety after the fact, with no premarket review of the finished product. An OTC drug is regulated through the monograph system, a standing rulebook for an entire category that a manufacturer must match exactly, without submitting its own proof to the FDA. A prescription drug goes through the process Chapter 22 describes in full: years of preclinical work, three phases of human trials, and an individualized FDA review of that drug's own data. The same active chemical can appear in more than one box at a different concentration or claim, benzoyl peroxide is an OTC acne drug at low strengths and a prescription-strength ingredient at higher ones, which is a big part of why cosmetic labeling confuses even attentive consumers.

From a fermentation tank to a store shelf

A finished tube of cream starts, like most manufactured goods, with raw ingredient sourcing, and cosmetic ingredients come from a wider mix of sources than most shoppers assume. Hyaluronic acid, ubiquitous in modern moisturizers for its ability to hold water in skin, used to be extracted from rooster combs; nearly all commercial supply today instead comes from bacterial fermentation, most commonly using Streptococcus equi subspecies zooepidemicus grown in steel tanks, where the bacteria are fed sugars and produce hyaluronic acid as part of their own cell structure, later harvested and purified to the molecular weight a given formula needs (lower weights for cosmetics, higher for medical uses like joint injections). Common preservatives have their own supply chains: phenoxyethanol, found in roughly a quarter of personal care products today, is a synthetic compound (occurring in trace amounts in green tea but produced industrially from petrochemical feedstocks) that became the default paraben substitute after "paraben-free" turned into a near-mandatory marketing claim around 2015. Parabens themselves, introduced to cosmetics, food, and pharmaceuticals in the 1930s, are synthesized alkyl esters of a naturally occurring acid and remain, despite their reputation, among the most thoroughly studied preservatives in use.

From there, ingredients move to a formulation and R&D lab, where chemists combine actives, emulsifiers, preservatives, and stabilizers into a prototype and run it through stability testing (does it separate, discolor, or change smell after weeks at high heat, or a freeze-thaw cycle simulating a truck outside a warehouse) before it's scaled up. Manufacturing happens on batch mixing and filling lines: jacketed tanks heat and blend the oil and water phases under controlled speed, the batch is tested against the approved formula, and filling lines dose the finished product into jars, tubes, or bottles at commercial speed, sealing and labeling each one.

Unlike drugs, where Current Good Manufacturing Practice (CGMP) has been a legally binding federal requirement for decades, cosmetics manufacturing in the U.S. operated for most of its history under voluntary guidance only, incorporating elements of ISO 22716, the international standard for cosmetic production and storage, without making either binding. That changed only partially with MoCRA, which directed the FDA to issue binding cosmetic GMP regulations, a rulemaking still in progress years after the law passed. Finished product then moves through the same wholesale and retail chain that carries almost every packaged consumer good to a shelf, minus the temperature-controlled handling a vaccine or biologic drug needs.

Who makes sure the cream works and doesn't grow anything

Cosmetic chemists and formulators, typically trained in chemistry or a related science, design the product: choosing the emulsifier system, balancing actives against stability, and iterating through dozens of lab batches before it's ready for manufacturing. Regulatory affairs specialists sit between the lab and the law, reviewing formulations and labeling against FDA rules, preparing the paperwork MoCRA now requires, and tracking changes that can force a reformulation with little notice. Clinical testing coordinators run the studies behind claims like "clinically tested" or "dermatologist tested," recruiting volunteer panels, often just a few dozen people, to use a product while researchers measure some outcome, hydration, wrinkle depth, self-reported irritation, that becomes the evidentiary basis, however thin, for the claim on the box. And quality control microbiologists run preservative efficacy testing, deliberately contaminating a sample with standardized doses of bacteria, yeast, and mold, then testing over the following days whether the preservative system suppresses it fast enough, the direct laboratory proof that a product won't grow something dangerous after months on a humid bathroom shelf.

Where this came from

For the first three decades of the 20th century, nothing at the federal level stopped a company from selling a cosmetic that was actively dangerous, as long as the label didn't lie about its contents. The 1906 Pure Food and Drug Act addressed food and drug purity and labeling but said essentially nothing about cosmetics as their own category. That gap became impossible to ignore because of a single product: Lash Lure, a permanent eyelash and eyebrow dye sold in the early 1930s, formulated around paraphenylenediamine (PPD), an aniline, coal-tar-derived dye. In a meaningful share of users, PPD triggered severe allergic reactions, and the eye area is especially vulnerable because the skin around the eyelids has the thinnest stratum corneum on the human body, the same barrier layer described earlier in this chapter. Documented cases from 1933 describe severe eyelid dermatitis, corneal ulceration, and blindness; one 52-year-old woman died in 1934 after having her eyebrows plucked before a Lash Lure application let a bacterial infection take hold. The FDA, lacking legal authority to pull the product from shelves, instead built a traveling exhibit nicknamed the "Chamber of Horrors" for the 1933 Chicago World's Fair, displaying before-and-after photographs of a woman blinded by the dye to make the case, to the public and to invited guests including Eleanor Roosevelt, that cosmetics needed federal oversight.

Congress moved slowly, and only after another, unrelated disaster, the 1937 elixir of sulfanilamide poisoning that killed roughly a hundred people, finally forced the issue. The Federal Food, Drug, and Cosmetic Act of 1938 was the result, giving the federal government authority over cosmetics as a defined legal category for the first time, prohibiting adulterated or misbranded products. Lash Lure was the first product seized under the new law, and the FDA soon moved to effectively ban PPD's use in eyelash and eyebrow dyes sold in the United States.

What followed was a strikingly long quiet period. Drug regulation kept evolving, most significantly with the 1962 Kefauver-Harris Amendment covered in the chapter on prescription medicine, but cosmetics regulation barely moved for the rest of the 20th century and into the 21st: no premarket approval, no mandatory recall authority, not even a requirement that manufacturers tell the agency what facilities they operated or what they sold. That gap lasted 84 years, until MoCRA closed part of it in 2022.

Standards and coordination

Two very different kinds of standard-setting cover this world, one built by industry, one only recently rebuilt by Congress.

The Cosmetic Ingredient Review (CIR), established in 1976 by what's now the Personal Care Products Council with FDA and Consumer Federation of America backing, is an industry-funded but procedurally independent panel of outside dermatologists, toxicologists, and pharmacologists who review published safety data on individual ingredients, with FDA and consumer advocacy representatives sitting in as non-voting observers. As of the most recent published tally, CIR had reviewed 4,740 ingredients since 1976, finding the large majority safe as used or safe with qualifications. CIR's findings are influential, cited constantly in regulatory literature, but aren't legally binding; a company can, in principle, ignore an unfavorable opinion and keep selling, though at real legal and reputational risk.

Real regulatory teeth arrived only in 2022, with MoCRA, signed into law on December 29, 2022, and the first major overhaul of U.S. cosmetics law since 1938. MoCRA gave the FDA, for the first time, mandatory recall authority: if the agency finds a reasonable probability that a cosmetic is adulterated or misbranded in a way that could cause serious harm, and the responsible company won't recall it voluntarily, the FDA can now order the recall itself. It created facility registration and product listing requirements, with enforcement beginning July 1, 2024, so the FDA now has a working list of who makes cosmetics in the U.S. and what they contain. It created a formal serious adverse event reporting duty, requiring the "responsible person" named on a label to report any serious adverse event within 15 business days. And it directed the FDA to write new rules on fragrance allergen labeling and standardized asbestos testing for talc-based products, both still working through rulemaking well past their original deadlines.

Don't be confused: a registered facility is not an approved product. MoCRA requires cosmetic facilities to register and products to be listed, but neither step involves the FDA reviewing or approving a formula or safety data before sale. It's closer to a census than a review: the FDA now knows a facility and product exist and can act if something goes wrong, but registering isn't the agency vouching for what's in the bottle.

The OTC monograph system underwent its own overhaul in 2020, when the CARES Act replaced the old notice-and-comment rulemaking process, which had made monograph updates take decades, with a faster administrative order process run through the FDA's Center for Drug Evaluation and Research. Sunscreen shows why that mattered: the FDA first called for sunscreen safety data in 1972, a 1999 final monograph was proposed and then stayed indefinitely, and it took until 2011 and 2018 to finalize the current broad spectrum and SPF testing requirements, decades after the review began, largely because the old rulemaking process was so slow.

Keeping it working

A cosmetic doesn't stop being tested once it leaves the factory. Preservative efficacy testing, often called a "challenge test," is the main ongoing tool. Under standards like the U.S. Pharmacopeia's USP <51> Antimicrobial Effectiveness Test, microbiologists deliberately spike a finished-product sample with standardized doses of five challenge organisms, commonly Staphylococcus aureus, Pseudomonas aeruginosa, E. coli, Candida albicans, and Aspergillus brasiliensis, then measure over 14 to 28 days whether the preservative system reduces that contamination by an accepted margin (commonly a 99.9 percent or greater reduction within 14 days, with no rebound growth by day 28). Products that fail get reformulated before they ship.

That testing supports a small icon many shoppers have seen but never quite understood: the period-after-opening (PAO) symbol, an open-jar icon printed with a number, as in "12M" or "24M." It originated in EU cosmetics regulation, which requires it on any product with a shelf life over 30 months as an alternative to a fixed expiration date, and has spread well beyond Europe because large manufacturers sell into the EU and reuse the same packaging globally. The number certifies how many months, after first opening, the manufacturer's stability and preservative data support the product remaining safe and effective under normal use, the printed output of the same challenge testing described above, applied to the period after a product's protective seal no longer applies.

When it breaks

Failures in this world fall into two very different buckets: contamination that slips past quality control, and harm from an ingredient nobody adequately tested in the first place.

Microbial contamination is the more common, immediately dangerous failure, common enough to have been studied systematically. A cross-sectional analysis of FDA recall data published in the Journal of the American Academy of Dermatology found 334 voluntary recalls of cosmetic dermatology products between 2011 and 2023, and microbial contamination was by far the leading cause, responsible for roughly three-quarters of them, with bacteria (most often Pseudomonas and Burkholderia species) the contaminant in the large majority of cases. These are exactly the failures preservative efficacy testing exists to prevent, and their persistence shows the system catches real problems after the fact rather than eliminating the risk entirely.

The other kind of failure is slower and far more contested: harm from an ingredient in wide use for decades before its risk became undeniable. Talc, used for its silky texture in baby powder and other cosmetics for most of the 20th century, can occur in natural deposits close to asbestos, a known human carcinogen, and plaintiffs in thousands of lawsuits have alleged that talc products, most prominently Johnson & Johnson's baby powder, were contaminated with asbestos and caused mesothelioma and ovarian cancer in long-term users. J&J has denied its products were unsafe, while juries have repeatedly disagreed at enormous scale, including a $1.56 billion Maryland verdict in December 2025, the largest individual award in the litigation so far; after three bankruptcy-based settlement strategies were rejected by courts, J&J said in 2025 it would litigate remaining cases individually instead. MoCRA directed the FDA to issue a standardized asbestos test for talc products, exactly the gap this litigation exposed, but the agency's proposed rule, published in December 2024, was withdrawn in November 2025 for further review, meaning that more than three years after MoCRA passed, the U.S. still has no standardized, legally mandated asbestos test for talc cosmetics.

Set against both is a subtler, everyday failure that generates no lawsuits and no recalls: marketing language that sounds like a regulatory claim but isn't one. Phrases like "dermatologist tested," "clinically proven," and "clean" have no standardized legal definition in U.S. cosmetics law. The FTC requires some substantiation behind any claim, but it need not be published, peer-reviewed, or rigorous; "dermatologist tested" can describe a large controlled study or one dermatologist glancing at a formula sheet, and a shopper has no way to tell which.

The scale of it

This is a genuinely enormous consumer market, even if estimates vary widely depending on what's being counted. Industry research firms put the global cosmetics market somewhere between 350 billion and over 600 billion dollars, depending on methodology and how broadly "cosmetics" is defined against the wider "beauty and personal care" category; Grand View Research separately estimated the U.S. cosmetics market alone at roughly 63 billion dollars as of 2023, growing over 6 percent a year. The narrower OTC drug segment, the acne treatments, dandruff shampoos, antacids, and sunscreens sold under the monograph system, is its own large market, estimated at somewhere between about 40 billion and 55 billion dollars in the U.S. alone for 2024. And underneath both categories, industry surveys suggest the average woman in the U.S. applies around a dozen personal care products containing over a hundred distinct ingredients to her body every single day, a volume of routine, largely unregulated exposure with no real equivalent anywhere else in consumer product law.

Trade-offs and what's next

MoCRA is, by any measure, the largest expansion of FDA cosmetics authority in 84 years, and also, by the account of the advocates who spent decades pushing for it, a partial one. The law still doesn't require independent premarket safety testing; recall authority and adverse event reporting are powerful but still reactive, catching harm only after a product is already on shelves and in use. The FDA, as of the mid-2020s, had formally restricted or banned only around a dozen substances in cosmetics, compared with more than 2,400 banned by the European Union, a gap groups like the Campaign for Safe Cosmetics point to as evidence that ingredient safety, not just facility oversight, remains the unresolved half of the reform. Critics also note that MoCRA's negotiated text restricted individual states' ability to pass tougher ingredient disclosure and safety substantiation laws, trading a federal floor for a lower state ceiling in states that had previously moved faster than Washington.

Consumer pressure is filling some of that gap from outside the regulatory system. "Clean beauty," an informal, marketing-driven movement rather than a legal category, pushes companies toward voluntary ingredient transparency and third-party certifications instead of the undefined phrases described above, precisely because no law requires any of it. Whether that market pressure substitutes for binding premarket safety review is an open argument, not a settled one; a product can carry every "clean" label a marketing team can print and still never have been tested against its ingredients' actual risk profile by anyone outside the company selling it.

Animal testing runs the opposite direction, toward less testing rather than more. The EU completed a full ban on animal testing for cosmetics, and on selling any cosmetic tested on animals anywhere in the world, in March 2013. The U.S. has no comparable federal ban; the FDA neither requires nor prohibits it, leaving the practical rules to a patchwork of state laws, starting with California's 2018 Cruelty-Free Cosmetics Act and since joined by roughly a dozen other states, that bar selling newly animal-tested cosmetics within their borders. That pressure, combined with the EU ban closing off a major export market, has pushed real investment into alternatives: more than 40 in vitro methods, including reconstructed human skin models grown from donated cells, are now accepted by international regulators as substitutes for live-animal testing in specific assessments.

Back to the bathroom counter

The night cream and the acne wash sitting side by side in your cabinet look like variations on the same idea: something in a tube, rubbed onto skin, promising some improvement. Legally, they aren't even close. One reached your shelf on the strength of its manufacturer's private confidence that it's safe, backed since 2022 by a facility registration and a recall power that didn't exist before. The other passed through a federal rulebook built specifically for its ingredient and its claim, tested under government- mandated methods before a company could put an SPF number or an acne claim on the label at all. Neither involves anything like the years of clinical trials a prescription drug goes through, and neither needs to, because neither is making the same kind of claim. What connects them both back to the earlier chapter on prescription medicine is the same idea running through this whole book: the label tells you almost nothing about the system standing behind it, and the testing, oversight, and legal exposure behind two products on the same shelf can differ enormously, invisibly, every time you reach for one.

The leap: what it replaced, and the work behind it

The quiet trust you place in a tube of cream is recent, and it was bought at a real price. For most of recorded history, putting a product on your skin was a genuine gamble. From the sixteenth century into the nineteenth, European aristocrats whitened their faces with Venetian ceruse, a paste of white lead, and paid for it with the classic signs of lead poisoning: hair loss, rotting teeth, scarred skin, and worse. Maria Gunning, Countess of Coventry, one of the celebrated beauties of her day, died in 1760 at 27, her death widely blamed on the very cosmetic she used to stay fashionable. The Lash Lure eyelash dye already described in this chapter belongs to the same lineage of products that could blind or kill the person applying them, and nobody was legally required to check first.

The most jolting case sits inside living memory. In the 1920s and 1930s, radioactivity was sold as a beauty ingredient: the Paris brand Tho-Radia put radium bromide into face powder and lipstick, promising firmer skin and fewer wrinkles, while a radium tonic called Radithor was drunk as a cure-all. The socialite and golfer Eben Byers died of radium poisoning in 1932 after consuming it for years, his jaw literally disintegrating. Meanwhile the young women hired to paint glowing radium dials, told the paint was harmless and instructed to point their brushes with their lips, were dying too: radium is estimated to have killed 112 dial painters. That was the world before regulation, a world where "it's on the shelf, so it must be safe" was simply false.

The leap is the difference between that world and the one where you can grab a cream without a second thought. An ordinary moisturizer today carries only 10 to 20 ingredients, but each is drawn from a global catalog of more than 16,000 named cosmetic ingredients, and the industry-funded Cosmetic Ingredient Review has published safety assessments on 4,740 of them since 1976. Behind the tube sit the regulatory affairs specialists, quality-control microbiologists, and independent labs this chapter has walked through, plus a federal rulebook that finally, in 2022, gave the FDA power to order a recall instead of merely asking. None of that protective machinery existed when Byers was drinking radium.

You live inside that machinery every day without noticing it. You smear on sunscreen, swallow an antacid, dab on a night cream, brush your teeth with a whitener, and you read almost none of the labels, because you don't have to: you trust, correctly most of the time, that someone checked. What that trust replaced was a countess dead at 27 and a factory of women with disintegrating jaws. The unremarkable tube in the cabinet is the visible tip of an invisible bargain, struck over a century of poisoned users, that lets you stop being your own safety inspector.

Real-world examples and recent developments

Behind the brand names on a bathroom shelf sit contract manufacturers, ingredient suppliers, and independent labs most shoppers never hear of, plus a regulatory landscape still actively changing.

  • Cosmax (founded 1992, South Korea): the world's largest cosmetics ODM (original design manufacturer), formulating and producing products later sold under other companies' brand names rather than its own. Its single Guangzhou plant can turn out roughly 400 million units a year, and the company now formulates and manufactures for more than 3,300 beauty brands worldwide, standing behind the "formulation and R&D lab" and "batch mixing and filling lines" described earlier in this chapter on nearly every label it fills.
  • Croda International (founded 1925, headquartered in Snaith, England): a specialty chemical company supplying more than 200 active ingredients to the cosmetics industry, including Matrixyl, a peptide ingredient that became close to an industry standard active in anti-aging creams. A concrete example of how many "actives" printed on a jar trace back to a chemical supplier most shoppers have never heard of, not the brand on the label.
  • Intercos (founded 1972, headquartered in Agrate Brianza, Italy): the leading contract manufacturer of color cosmetics worldwide, running 16 production plants and 11 R&D centers and formulating lipstick, pressed powder, mascara, and eyeliner for 24 of the world's 30 largest beauty companies. Like Cosmax, its name almost never appears on a finished package.
  • Valisure (2024): an independent pharmacy and analytical testing lab based in New Haven, Connecticut, that filed an FDA citizen petition in March 2024 showing benzoyl peroxide, the active ingredient in many OTC acne treatments, can chemically break down into benzene, a known carcinogen, at levels its own testing measured as high as hundreds of times the FDA's conditional limit. Valisure operates outside the manufacturers, retailers, and government labs that normally do this kind of testing, and its petitions have prompted several FDA reviews and recalls in recent years.

Recent developments

  • California's PFAS-Free Cosmetics Act (AB 2771): signed into law in September 2022 and taking effect January 1, 2025, it bans more than 12,000 individual PFAS ("forever chemical") compounds from cosmetics sold in California when intentionally added, a state-level rule reaching further than the federal restrictions MoCRA left in place.
  • FDA acne product recalls following Valisure's benzene findings (2024): after Valisure's petition, the FDA tested 95 benzoyl peroxide acne products and found six with elevated benzene levels, leading to voluntary recalls of specific lots including La Roche-Posay Effaclar Duo, Proactiv Emergency Blemish Relief Cream, and Walgreens branded acne treatments, while the agency concluded the added cancer risk from those particular levels was very low.

Glossary

Emulsion. A stable mixture of two normally immiscible liquids, typically oil and water, held together by an emulsifier so one liquid disperses evenly through the other instead of separating.

Emulsifier. A molecule with both oil-attracting and water-attracting ends that coats droplets of one liquid phase to keep an emulsion from separating.

Stratum corneum. The outermost layer of skin, made of tightly packed dead cells bound by lipids, that forms the primary barrier controlling what can pass into or out of the body through the skin.

Intended use. The legal standard, based on a product's label and marketing claims rather than its ingredients, that determines whether it is regulated as a cosmetic or a drug under U.S. law.

OTC monograph. A standing FDA rulebook for an entire category of over-the-counter drug, listing ingredients and concentrations recognized as safe and effective, which manufacturers can follow without submitting an individual product for FDA review.

Premarket approval. Formal regulatory review and sign-off required before a product can be sold; cosmetics historically required none, while prescription drugs require extensive premarket review.

Modernization of Cosmetics Regulation Act (MoCRA). The 2022 U.S. law that gave the FDA mandatory recall authority, facility registration, product listing, and adverse event reporting requirements for cosmetics, the first major overhaul of federal cosmetics law since 1938.

Cosmetic Ingredient Review (CIR). An industry-funded but procedurally independent expert panel, established in 1976, that publishes safety assessments of individual cosmetic ingredients; influential but not legally binding.

Preservative efficacy testing (challenge test). A laboratory test, such as USP <51>, in which a product is deliberately contaminated with standardized microorganisms to measure whether its preservative system suppresses growth quickly and durably enough to meet an accepted standard.

Period-after-opening (PAO) symbol. An open-jar icon with a number of months, originating in EU cosmetics regulation, certifying how long a product remains safe and effective for use after it is first opened, based on stability and preservative testing.

Current Good Manufacturing Practice (CGMP). Legally binding manufacturing quality standards required for drugs; cosmetics manufacturing in the U.S. has historically operated under voluntary, non-binding guidance instead.

Responsible person. Under MoCRA, the manufacturer, packer, or distributor named on a cosmetic product's label, legally required to report serious adverse events to the FDA within 15 business days.

Bacterial fermentation (cosmetic ingredient production). A manufacturing method using bacteria grown in controlled tanks to produce ingredients like hyaluronic acid, replacing older animal-derived extraction methods.

Sources and notes

Open questions

  • Estimates of the global and U.S. cosmetics and OTC drug market vary substantially between research firms depending on how the category is defined; treat the figures here as representative ranges, not settled totals.
  • The FDA's rulemaking on fragrance allergen labeling and standardized talc asbestos testing, both required by MoCRA, remained unfinished and past their original statutory deadlines as of the time of writing, so the exact final requirements are still unresolved.
  • Whether U.S. state-level cruelty-free laws and industry adoption of in vitro alternatives will eventually add up to a de facto national standard, without a federal law forcing it, is still an open, evolving question.

That's where this book's ninth part, and its search for what else hides in plain sight, comes to a close. Clothing, cosmetics, and the other everyday categories this part examined turned out to hide the same basic pattern the rest of the book already traced through water, power, food, and medicine: a plain object on a shelf, sitting quietly on top of chemistry, law, labor, and history that most of us never have reason to ask about until something goes wrong. If any of it now reads differently than it did before you started, the best test of that is to go back to the very first page and see whether a flush of a toilet still looks as simple as it used to. What happens after you flush a toilet? 👉