Is Surrogacy Ethical? Europe Bans Wombs, Not Brothels
Europe bans commercial surrogacy on dignity grounds while licensing paid sex work and paid plasma donation. Germany, Austria and Hungary do all three at once. The line is a taboo, not a principle.
Elena Marsh covers science policy for Stanford Tech Review, with a focus on federal research funding, biosecurity oversight, and how Washington governs the labs it pays for.

In the October 2026 issue of The Atlantic, Helen Lewis publishes a long, careful and frequently devastating piece of reporting on the American commercial surrogacy industry, titled "Mothers for Sale." It contains everything a reader could want from magazine journalism: named sources, contract language, court records, a fertility conference rendered in merciless detail down to the sperm-shaped stress balls.
The reporting is sound. The premise underneath it is not.
That premise is stated most plainly not by Lewis but by the authorities she cites approvingly. A 2025 United Nations report declares that commercial surrogacy "constitutes the sale of children, which is a crime," and that such arrangements "can amount to or resemble slavery." Emmanuel Macron calls the practice "not compatible with the dignity of women." Every member state of the European Union has banned it. The shared assumption is that paying a woman to gestate is categorically unlike paying a person to do anything else — that money entering this particular transaction converts a service into a sale, and a woman into merchandise.
It does not. And the countries making the argument have already conceded the point in their own statute books.
There is no such thing as disembodied labor
Start with the general claim, because the surrogacy debate is a special case of it.
Every job anyone has ever been paid for is the rental of a body. The bricklayer sells his back and will feel it at fifty. The long-haul driver sells his spine, his circadian rhythm and a measurable slice of his life expectancy. The violinist sells tendons that will eventually fail her. The line cook sells forearms that scar. The programmer sells eyes, wrists, and the finite quickness of a brain that will not be as fast at sixty as it was at thirty. A nurse sells lifting strength and the emotional regulation to be shouted at by frightened people for twelve hours. A soldier sells the whole thing at once.
Hands, feet, or brain: pick any occupation and it resolves into some part of a human being, spent, permanently, for money. The distinction between "selling your labor" and "selling your body" that carries the entire weight of the anti-surrogacy argument does not survive contact with any actual labor market. What we call labor is the body, metered out.
The usual reply is that gestation is different because it is dangerous, and because the danger is borne by a woman who needs the money. Both halves are true. Neither is distinguishing.

American pregnancy killed 17.9 women per 100,000 live births in 2024, according to the CDC's National Center for Health Statistics — 649 deaths. American farming, fishing and forestry work killed 24.4 per 100,000 full-time-equivalent workers in 2023, according to the Bureau of Labor Statistics; transportation, 13.6; construction and extraction, 12.9. The denominators are not identical — an FTE-year of work against one completed pregnancy of roughly nine months — so treat this as an order-of-magnitude comparison rather than an exact one. It is still the relevant order of magnitude. Carrying a pregnancy for money is roughly as dangerous as commercial fishing and rather more dangerous than driving a truck, and no European government proposes to ban either.
Nor is the risk premium hidden. A 2024 cohort study of Ontario births published in Annals of Internal Medicine found severe maternal morbidity in 7.8% of gestational-carrier pregnancies against 4.3% for IVF and 2.3% for unassisted conception — the three-times figure Lewis cites. That is a real finding and it deserves a real response. The response that every other dangerous trade receives is mandatory insurance, enforced safety standards, an independent medical advocate, and a licensing regime. The response gestation receives is prohibition.
The line Europe actually draws
Here is the part of the argument that the dignity framing cannot absorb.
Europe has not banned the sale of bodily use. It has banned the sale of one organ, while licensing, taxing and regulating the sale of the organs on either side of it.
| Jurisdiction | Paid sexual services | Compensated plasma donation | Commercial surrogacy |
|---|---|---|---|
| Germany | Legal, licensed, taxed | Permitted | Prohibited |
| Austria | Legal, registered, health-checked | Permitted | Prohibited |
| Hungary | Legal, regulated | Permitted | Prohibited |
| Czechia | Not criminalized | Permitted | No legal framework |
| Netherlands | Legal, licensed since 2000 | Unpaid only | Commercial arrangements prohibited |
| Switzerland | Legal, regulated since 1942 | Unpaid only | Prohibited by constitution |
| Greece | Legal in state-licensed premises | Unpaid only | Altruistic only, court pre-approval |
| France | Selling legal, buying criminalized (2016) | Unpaid only | Contracts void (Civil Code art. 16-7) |
| United States | Prohibited outside parts of Nevada | Permitted, paid | Legal in most states |
Sources: European Parliament study on member-state prostitution regulation and national statute; EUobserver on the EU plasma regime; Article 16-7 of the French Civil Code.
Three European countries — Germany, Austria and Hungary — simultaneously license the paid use of the vagina, license the paid extraction of plasma, and criminalize the paid use of the uterus. That is the finding, and it is the whole argument in one line. Method: legal status of three forms of compensated bodily labor, compiled from national statute across nine jurisdictions in the table above; three satisfy all three conditions.
Put it as bluntly as the facts allow. In Munich, a woman may lawfully be paid to let a stranger use her vagina. She may lawfully be paid to let a machine draw plasma out of her arm twice a week. She may not lawfully be paid to let an embryo use her uterus. The organs are centimeters apart. The dignity that is violated by the third transaction is apparently untouched by the first two.
If the objection is intimacy, sex work is more intimate. If the objection is bodily risk, plasmapheresis is trivial and repeated commercial sex is not. If the objection is that money corrupts consent where economic need is present, that argument applies to the licensed brothel with more force, not less. If the objection is commodification of a bodily capacity, both of the permitted trades are exactly that. Every principle offered against paid gestation strikes the two transactions Europe already permits, and strikes them harder.
The plasma case is the most awkward, because Europe does not merely tolerate it — it depends on it. Compensated plasma donation is permitted in four EU states: Austria, Czechia, Germany and Hungary. Those four supply close to half of all plasma collected in the Union, and the EU imports a large further share of its plasma-derived medicines from paid American donors. The continent that calls payment for bodily contribution an affront to human dignity treats its own citizens' immunoglobulin supply as an exception, and buys the shortfall from a country whose donors it considers exploited.
France pays €1,900 to risk your brain
The clearest single case sits in France, the jurisdiction whose Civil Code declares every gestational agreement void as a matter of public order.
In January 2016, at the Biotrial facility in Rennes, healthy volunteers were paid to take an experimental FAAH inhibitor called BIA 10-2474 in a first-in-human trial. Participants were to receive €1,900, including travel expenses, for two weeks residence and dosing. Five men in the high-dose cohort developed hemorrhagic and necrotic brain lesions. Guillaume Molinet, forty-nine, an artist and father of four, died.
This was a lawful, regulated, ethics-committee-approved transaction in a country that holds paid gestation to be incompatible with the dignity of women. A French woman may accept €1,900 to let an unknown molecule cross her blood-brain barrier. She may not accept any sum to carry a pregnancy. Whatever principle produces that pair of rules, dignity is not it.
What the ban actually accomplishes
Prohibition has not removed the transaction. It has relocated it, which is a different thing, and which Lewis's own reporting documents better than any critic could.
Her piece follows the trade to Georgia, to Ukraine, to Northern Cyprus, to Mexico, and to Ghana, where an agency director cheerfully describes carriers housed in a dormitory beside the hospital under twenty-four-hour monitoring so they "can't drink, they can't smoke, they can't do drugs." Italy has responded by making it a crime punishable by up to a million euros for a citizen to return home with a child born to a surrogate abroad — a law that reaches across borders to punish the parent and does precisely nothing for the Georgian or Ghanaian woman it is nominally protecting.
That dormitory in Accra is not what prohibition prevents. It is what prohibition produces. Demand for gestation is inelastic in the way that demand for a child is inelastic, and a banned market does not clear at zero; it clears in whichever jurisdiction has the least capacity to protect the person taking the risk. This is the same result the same policy produces for every other prohibited service, and there is no reason to expect an exception here.
Every problem in the reporting is a labor-standards problem
Read Lewis's catalogue of American failures again with the word worker substituted, and it reads like a routine occupational-safety indictment:
- Contracts drafted entirely for the counterparty, with no guaranteed independent counsel for the carrier. Answer: mandatory independent representation at the commissioning party's expense, as in union certification or real-estate closings.
- No cap on transferred embryos, because no rule requires single-embryo transfer. Answer: a clinical standard with a license attached.
- NDAs that prevent a carrier from describing her own medical treatment. Answer: these are already void in most employment contexts; make them void here.
- Contract terms that bind a carrier's residence for months. Answer: unenforceable restraint; strike the clause, keep the contract.
- No medical or criminal disclosure required of the commissioning parents, while carriers are screened exhaustively. Answer: symmetric disclosure, which is Lewis's own recommendation.
- No insurance requirement, leaving carriers with medical debt when the intended parents disappear. Answer: mandatory disability and life cover, priced into the fee, as with any hazardous trade.
Every item on that list is a standard the United States already imposes on ordinary employment and has simply failed to impose here. None of them requires believing that gestation is not work. All of them require believing that it is.
Where the analogy genuinely stops
One objection survives, and it is the serious one. It is not about the woman.
A labor contract binds two consenting parties. This one produces a third who consented to nothing and whose entire welfare is determined by the screening applied to the purchaser. That is not a market objection; it is a child-protection objection, and on that ground American practice is indefensible. The Arcadia case Lewis reports — a couple accumulating twenty-one children through an agency they had incorporated for the purpose, with no charges filed because none of it was illegal — is not an argument against paying carriers. It is an argument that a country conducting rigorous fit-and-proper screening on adoptive parents conducts none at all on commissioning parents.
Fix that with the tool built for it. Age limits, criminal-record checks, caps on commissioned children, and a home study, applied to the buyer exactly as adoption applies them. Those are licensing requirements on the party acquiring the child. They have nothing to do with whether the woman gets paid, and prohibiting her payment does not deliver a single one of them.
America does not need a European chaperone
Lewis observes that Europeans regard American enthusiasm for paid surrogacy the way they regard American guns and American executions: inexplicable. The comparison is offered as an argument. It is not one. It is a report of a mood.
A moral consensus is not a moral proof, and the consensus in question belongs to a set of governments that license brothels, sell their citizens' plasma into a continental shortage, pay healthy volunteers to take molecules that occasionally kill them, and answer the resulting surrogacy demand by driving it to Tbilisi and Accra where nobody has to look at it. That is not a superior ethical framework. It is a different distribution of squeamishness, arrived at by different historical accident, and enforced with the confidence of people who have never been asked to reconcile the three columns of that table.
The United States has its own inconsistency, and it is the mirror image: it permits paid gestation and criminalizes paid sex nearly everywhere. Neither continent is operating from a principle. Both are operating from a taboo, and each finds the other's taboo unaccountable, which is what taboos are for.
But a country is entitled to work out its own answer, in its own legislatures, against its own convictions — and to be judged on the result rather than on the distance between its answer and Brussels'. The American answer is currently a bad one, for a reason that has nothing to do with payment: it is unregulated. The correction is a labor and licensing regime, written by Americans, that treats gestational carriage as what it plainly is — dangerous, skilled, compensated work performed with the body, like every other kind.
Everyone who has ever been paid has been paid for some part of themselves. The women in Lewis's reporting were not degraded by being paid. They were degraded by being paid without a lawyer, without insurance, without a cap on the risk transferred to them, and without anyone checking who was on the other side of the contract. That is a failure of regulation, and prohibition is what a government reaches for when it would rather not do the regulating.
Cover image: generated illustration. Chart by Stanford Tech Review from BLS and CDC/NCHS data.