Psychology › Thinking, memory & biases › Nudges and choice architecture
Defaults and 401(k) auto-enrollment
2001· Finding or experiment
Well replicatedStrong, repeated evidence.
People tend to stick with whatever happens if they do nothing, such as automatic enrollment in a 401(k) retirement plan.
See it on the Psychology map →What it says
In 2001 economists Brigitte Madrian and Dennis Shea studied a large US company that began enrolling new hires in its 401(k) automatically unless they opted out. Participation among new employees jumped from about 37% to 86%. But most also kept the default 3% contribution and ultra-safe money-market fund, saving less than many would have chosen. Defaults work elsewhere too: countries where organ donation is opt-out have far higher registration than opt-in ones.
Why it matters
Defaults cost almost nothing yet shift behavior massively. US law now builds them in: the 2006 Pension Protection Act encouraged auto-enrollment, and SECURE 2.0 (2022) requires it for most new 401(k) plans from 2025.
Think of it like
Like an evergreen lease clause: whatever happens if nobody acts is what usually happens.
The evidence
Status: well replicated. Payroll records in Madrian and Shea's study and many later employer studies; Johnson and Goldstein's 2003 comparison of European organ-donor rates; meta-analyses rank defaults among the strongest nudges.
Open questions
Why defaults work so well: inertia, the implied endorsement of whoever set them, or avoiding a hard choice. Probably all three, in proportions that vary by setting.
Connections (2)
▸Defaults and 401(k) auto-enrollment are the strongest example of Nudges and choice architectureWell replicated
The connection
Of all nudges, setting the default usually has the biggest and most reliable effect. Enrolling workers in a 401(k) automatically, unless they opt out, more than doubled participation in Madrian and Shea's study. Defaults also work for organ donation, green-energy plans and privacy settings, which is why they anchor the nudge toolkit.
Who showed it and when
Brigitte Madrian and Dennis Shea studied a large US company that switched to auto-enrollment in 1998; they published in 2001.
The evidence
A 2022 meta-analysis by Stephanie Mertens and colleagues found decision-structure nudges such as defaults had the largest effects; the 2006 and 2022 US retirement laws built them in.
Open questions
Whether defaults change real outcomes or mostly paperwork; opt-out organ-donor laws, for example, raise registrations more reliably than actual donations.
▸Defaults and 401(k) auto-enrollment counteract Present bias (hyperbolic discounting)Well replicated
The connection
Present bias makes people put off costs now for benefits later: saving can always start next month. A default flips the effort, so saving happens unless you act to stop it. Thaler and Shlomo Benartzi's Save More Tomorrow went further, committing workers in advance to raise contributions at future pay raises, when the pain feels distant.
Who showed it and when
Save More Tomorrow was first tried at a midsized US manufacturer from 1998; Thaler and Benartzi published the results in 2004.
The evidence
At that firm, participants' average saving rate rose from 3.5% to 13.6% over about 40 months, and most stayed in the program.
Open questions
Whether defaults truly fix present bias or simply exploit inertia; people who never actively chose may not understand or value their own savings.