Defaults, and the organ donation figures. · published 3 September 2026
The default is genuinely one of the strongest levers in behavioural science — it doubled consent even in a randomised online experiment — but the consent gap is not the donation gap: presumed-consent countries run only roughly a quarter to a third higher in actual donations, and Spain, the country everyone cites as proof, sat unremarkable for a decade after its 1979 law and only moved once it built full-time hospital donor coordinators, which means the paperwork changed the answer while the staffing changed the outcome.
These are the working notes, unedited. Everything the video
states as fact comes from here, and nothing else may be stated as fact.
The sections on limits and on what the finding does not show are
part of the brief, not an afterthought — see
the method.
Eric Johnson and Daniel Goldstein, “Do Defaults Save Lives?”, Science, 2003.
What they compared
European countries fall into two groups. In opt-in countries you are not a donor unless you register as one. In opt-out, or presumed consent, countries you are a donor unless you register that you are not. The forms differ; the question does not.
The gap is enormous. In their figures, effective consent in opt-out countries sat in the high nineties — Austria at 99.98 per cent — while opt-in countries were in the teens or lower, with Germany around 12 per cent and Denmark around 4. Austria and Germany are the pair worth naming: neighbours, closely comparable in language, wealth and medical systems, separated by which box the form leaves ticked.
They also ran a controlled online experiment, randomly assigning people to an opt-in, opt-out or neutral form. Consent was roughly twice as high when the default was to be a donor. That matters, because it rules out the objection that these countries simply differ in how people feel about donation.
Why defaults work
Changing one takes effort. It also reads as a recommendation — the default looks like what a sensible person does, or what the institution expects. And whichever option you start from tends to feel like the one you own.
The part that gets left out. Consent on paper is not an organ in an operating theatre. The difference in actual donation rates between the two systems is real but far smaller than the consent gap suggests — Abadie and Gay (Journal of Health Economics, 2006) estimated presumed consent countries run roughly a quarter to a third higher, once other factors are controlled for. Not 99 against 12. Families are still consulted in practice, and a registered donor whose family refuses is usually not a donor.
Spain is the country most often produced as proof that presumed consent works, and it is the weakest example available. Spain has had a presumed consent law since 1979, and its donation rate was unremarkable for the decade that followed. What changed was the creation of the national transplant organisation in 1989 — full-time donor coordinators in hospitals, intensive care staff trained to identify potential donors, and families approached by people who do that work for a living. The law was already there. The infrastructure was the variable.