Default effects: what happens when the preset option decides for you
Madrian and Shea studied one large US company and found that at 3 to 15 months of tenure, 37% of employees hired just before automatic enrollment participated in the 401(k), versus 86% of those hired under it. The catch: 76% of the auto-enrolled participants stayed at the 3% default contribution rate and 80% of their contributions went to the default money market fund.
The natural experiment
Madrian and Shea analyzed 401(k) behavior at a large US corporation before and after it switched from requiring employees to affirmatively elect participation to enrolling them automatically unless they opted out. None of the economic features of the plan changed. Only the default did. That is what makes the study useful: match rates, fund menus and eligibility were held constant while the preset answer flipped, so the difference in behavior cannot be attributed to the plan becoming more generous. The authors describe the result as evidence that large changes in savings behavior can be motivated by what they call the power of suggestion.
The size of the effect
Comparing employees at the same point in tenure is what makes the comparison fair, since participation rises steeply with time at a company. At 3 to 15 months of tenure, the cohort hired just before the switch participated at 37 percent, while the cohort hired under automatic enrollment participated at 86 percent. The paper notes that participation under automatic enrollment even exceeded that of much longer-tenured employees hired under the old regime, and that automatic enrollment sharply compressed the differences in participation across demographic groups, with the largest gains among the groups that had participated least before.
The caveat that matters more than the headline
The same paper reports that the default did not only decide whether people participated, but also how. Among participants in the automatically enrolled cohort, 76 percent contributed at the default rate of 3 percent, and 80 percent of that cohort's contributions went to the default money market fund, even though very few employees hired before automatic enrollment had chosen that particular combination. The authors attribute this to a mix of inertia and employees reading the default as advice from the company. So the honest summary is not that defaults help people. It is that defaults decide, in both directions, and whoever sets them is making the decision.
The same pattern outside retirement
Johnson and Goldstein reported in Science in 2003 that the same mechanism shows up in organ donation. In an online experiment with 161 respondents, effective consent was about 42 percent when people had to opt in, about 82 percent when they had to opt out, and about 79 percent with no default at all, with only the opt-in condition significantly lower. Across European countries, the opt-in and opt-out groups had no overlap at all in effective consent rates, and the authors' own regression on actual cadaveric donation rates found a 16.3 percent increase in donation under an opt-out default. They also note registered consent does not always translate into a donation.
Why the effect is so large and so lazy
Johnson and Goldstein propose three reasons defaults work. People may read the default as a recommendation from whoever set it. Making an active decision takes effort, and accepting the preset takes none. And the default usually represents the status quo, so departing from it feels like a loss. None of those reasons involves the person having strong preferences about the outcome. That is precisely why the effect is powerful and why it should make you uneasy: it moves behavior most in the cases where people have thought about it least.
What this means for envelopes
If a preset is going to decide most of the outcome, the design question is who writes the preset. Left alone, your defaults are written by whoever is nearest your money: a subscription that renews, a card that is stored, a checking balance that reads as available. A saved budget template with recurring fills is the same mechanism turned around. You write the preset once, on a calm day, and then inertia works for the allocation instead of against it. Note the limit honestly: no study cited here tested a budgeting app, and defaults you set yourself are easier to override than defaults set by an employer.
Sources
Every source below was retrieved and checked. Findings are reported as the source states them.
-
The Power of Suggestion: Inertia in 401(k) Participation and Savings Behavior
— Quarterly Journal of Economics (figures read from NBER Working Paper 7682), 2001
At 3 to 15 months of tenure, 401(k) participation was 37% for the cohort hired just before automatic enrollment versus 86% for the cohort hired under it. Among the automatically enrolled cohort, 76% of participants contributed at the 3% default rate and 80% of contributions went to the default money market fund. Overall participation before automatic enrollment was 61%. -
Do Defaults Save Lives?
— Science, 2003
In an online experiment with 161 respondents, effective consent was roughly 42% under opt-in, 82% under opt-out and 79% with no default, with only opt-in significantly lower. Effective consent rates across European opt-in and opt-out countries showed no overlap, and a regression on actual cadaveric donation rates found a 16.3% increase under an opt-out default.
Common questions
Does automatic enrollment mean people saved more overall?
Not automatically. Participation rose sharply, from 37 to 86 percent at matched tenure, but the same study found most auto-enrolled participants remained at the 3 percent default contribution rate and in the default money market fund. A person who would have chosen 6 percent on their own but is nudged into participating at 3 percent is saving more than nothing and less than they might have. That is exactly why the paper treats the choice of default value as consequential, not just the choice to have one.
Is the organ donation finding about donation rates or paperwork?
Both, and they differ in size. The very large differences, with no overlap between opt-in and opt-out countries, are in effective consent, meaning registration status. When Johnson and Goldstein regressed actual cadaveric donation rates on the default, they found a smaller but still significant 16.3 percent increase. They explicitly list reasons registered consent does not always produce a donation, including family objections and medical mismatch.
Can I use this on myself, or does it only work when someone else sets the default?
It is weaker on yourself, and the research does not measure the self-set case. The employer default in Madrian and Shea required an affirmative action to escape, which is why inertia held. A default you wrote can be changed as easily as it was made. The realistic claim is smaller: a preset allocation removes the need to make the same decision repeatedly, which is worth something even when overriding it is easy.
Do these findings transfer to everyday spending?
Carefully, at best. Both studies are about one-time enrollment decisions with high stakes and low frequency, not about hundreds of small purchases a month. What transfers is the mechanism, that presets and status quo carry a lot of weight when attention is scarce. What does not transfer is any specific number. Nobody has shown that a default budget allocation moves grocery spending by anything like 37 to 86 percentage points, and no honest reading of these papers would imply it.
Run this budget on your phone
Envelope Budget puts these envelopes in your pocket. Assign every amount, log spending as it happens, and see what is actually left.
Get Envelope BudgetiPhone · manual entry, no bank connection · 7-day free trial