Why You Never Change the Setting
The pre-ticked box is the most powerful thing on the page, and the evidence for it has just been through the roughest audit the field has seen. Here is what survived, what did not, and what to ask of any setting you have never touched.
Key takeaways
- Brigitte Madrian and Dennis Shea found that switching one company's pension scheme to automatic enrolment raised participation sharply, and that most people then stayed on the exact default contribution rate and fund.
- Their second explanation matters more than the famous one: employees appeared to read the default as advice, not merely as the easy path.
- Gabriel Carroll, James Choi, David Laibson and Brigitte Madrian showed a third option beats both. Forcing an explicit choice raised enrolment by twenty-eight percentage points against opt-in.
- The organ donation evidence is real but not one-way. Opt-out countries have more deceased donors and fewer living ones, with total kidneys still higher.
- In 2022 two reanalyses cut the wider nudge literature down hard. Interventions that restructure the choice, the family defaults belong to, came out undecided rather than refuted, which is why we think the defaults evidence survives and the generalisation does not.
In 2001 two economists reported what happened when one American company stopped asking new employees whether they wanted a pension and simply enrolled them, with the option to leave. Participation rose sharply. The plan itself had not changed at all. Only the box that was already ticked.
In 2003 a paper in Science asked whether the same trick worked for organ donation, comparing countries where you have to opt in against countries where you have to opt out. In 2008 Nudge turned the idea into a movement, and governments hired behavioural teams. In 2021 a meta-analysis in PNAS reported that this whole family of interventions works, at a moderate average size across hundreds of studies. In July 2022 two letters in the same journal said that once you correct for publication bias, most of that evidence disappears.
That argument is still live. Meanwhile the setting you have never changed is deciding things about your money, your data and, in some countries, your organs.
## What the pension plans showed
Brigitte Madrian and Dennis Shea had access to something better than a laboratory: the administrative records of a company before and after it switched to automatic enrolment. Participation was significantly higher under the new arrangement, and a large proportion of the people enrolled automatically stayed on the exact default contribution rate and the default fund, even though almost nobody hired under the old system had chosen that particular combination.
Their explanation has two parts and both are worth keeping. The first is inertia, which is what everybody remembers. The second is that employees appeared to read the default as advice, as if the company had thought about it and this was the recommendation.
A default is the [easy path](https://thewhyweproject.com/articles/what-is-status-quo-bias) and also a message about what a reasonable person does, which is what separates it from the rest of the behavioural toolkit. Sticking with whatever you were given is the pattern in What is status quo bias.
Gabriel Carroll, James Choi, David Laibson and Brigitte Madrian later tested a third option that nobody talks about: [forcing new hires to make an explicit choice either way](https://thewhyweproject.com/articles/can-a-forcing-function-stop-mistakes), with no default at all. Requiring an active decision raised initial enrolment by twenty-eight percentage points against the standard opt-in procedure, producing a savings distribution after three months that would have taken thirty months to reach otherwise. Their model sets out when each regime wins. Active choice is best when people [procrastinate](https://thewhyweproject.com/articles/why-do-we-procrastinate) heavily and want very different things from each other. Automatic enrolment is better when people do not understand the decision they are being asked to make.
## What the organ donation paper claimed
A default is the easy path and also a message about what a reasonable person does.
Eric Johnson and Daniel Goldstein's 2003 policy forum is the version of this idea that escaped into public life. Using a mixture of national data and experiments, they reported that the consent default makes a large difference to how many people are registered donors, with opt-out countries far ahead.