Self Lab

Evidence-based self-improvement. No mechanism, no article.

Esc or click outside to close

Money & Growth · Mechanism

Save More Tomorrow: The Trick That Nearly Quadrupled Savings

Save More Tomorrow raised savings rates from 3.5% to 13.6% in 40 months. Here's the present-bias trick behind it, plus the honest caveat nobody mentions.

Alex by Alex July 12, 2026 12 min read
A staircase built from stacked coins climbing from left to right, each step a little taller than the last.

In 1998, a midsize manufacturing company handed two economists the keys to its retirement plan. Nobody got a lecture on discipline. Nobody downloaded a budgeting app. Workers simply agreed, months in advance, to save more later. Forty months on, their average savings rate had nearly quadrupled: from 3.5% of pay to 13.6% (Thaler & Benartzi, 2004).

The plan was called Save More Tomorrow, and the trick behind it isn’t really about money. It’s about a bug in how your brain prices the future. Psychologists and economists call it present bias.

This article runs on one premise: motivation is a mood, not a fuel tank. You’ll see how the trick works, the myth it buries, and the honest caveat most write-ups skip. Then you’ll point the same mechanism at any automatic transfer. This is psychology education, not financial advice.

The Bottom Line

  • Save More Tomorrow lifted average savings rates from 3.5% to 13.6% of pay in 40 months (Thaler & Benartzi, 2004).
  • The mechanism is present bias: sacrifices scheduled for later feel cheap, so schedule them for later.
  • Tie increases to raises and take-home pay never drops, so loss aversion never fires.
  • Mind the catch: defaults can anchor you low, and the career-long effect is smaller than the headline.

In this guide:

What is Save More Tomorrow?

Save More Tomorrow, or SMarT, is a savings-plan design from economists Richard Thaler and Shlomo Benartzi. Workers pre-commit to raising their savings rate each time they get a raise. In its first real-world test, the average rate climbed from 3.5% of pay to 13.6% in 40 months (Thaler & Benartzi, 2004).

The design has three moves, and every one targets a specific mental flinch:

  1. Commit early. Workers signed up about three months before their next raise. Agreeing costs nothing today, so agreeing is easy.
  2. Start after the raise. The increase begins with the first post-raise paycheck. Take-home pay never goes down, so the brain never registers a loss.
  3. Keep the exit. Anyone can opt out anytime. An open door makes walking in feel safe.

The results were strange by financial-education standards. 78% of the workers offered the plan joined. 80% of joiners stayed through four consecutive raises, stepping from 3.5% to 6.5%, 9.4%, 11.6%, and finally 13.6% (Thaler & Benartzi, 2004). Even the dropouts kept their already-higher rate. They just stopped climbing.

Notice the shape of the move. A raise is an anchor: a scheduled event you attach a behavior to. It’s habit stacking with a payroll calendar.

The Save More Tomorrow climb Average savings rate of Save More Tomorrow participants stepped from 3.5% of pay before the program to 6.5% after the first raise, 9.4% after the second, 11.6% after the third, and 13.6% after the fourth, across 40 months, per Thaler and Benartzi, 2004. The Save More Tomorrow climb Average savings rate, % of pay, over 40 months and four raises 15% 10% 5% 0% 3.5% 6.5% 9.4% 11.6% 13.6% start raise 1 raise 2 raise 3 raise 4 Source: Thaler & Benartzi (2004), Journal of Political Economy
Nobody found extra willpower. The schedule did the saving. Source: Thaler & Benartzi (2004).

Why does saving later feel easier than saving now?

Because your brain prices the future on a curve. Economists call the bug present bias, and David Laibson formalized the math as hyperbolic discounting in 1997 (Laibson, 1997). It’s why 78% of workers who wouldn’t save more today happily agreed to save more in three months (Thaler & Benartzi, 2004).

Here’s the curve in one choice. Laibson’s math predicts a flip. Offered $100 today or $110 next week, you grab the $100. Offered $100 in 52 weeks or $110 in 53 weeks, you wait. Same $10, same one-week delay. The only difference is distance. Up close, the discount curve is brutally steep. Far away, it goes flat.

Present bias has a sneaky second act. In 1999, Ted O’Donoghue and Matthew Rabin showed that people who assume they’ll “do it later” systematically over-delay, because later keeps moving (O’Donoghue & Rabin, 1999). You treat future-you as a different person: calmer, richer, mysteriously more disciplined. why your brain ignores future-you

SMarT doesn’t fight this bias. It hires it. The commitment lands in the flat part of the curve, where saving feels cheap. And by starting each increase with a post-raise paycheck, the number on your pay stub never shrinks. That flinch has a name: loss aversion, mapped by Kahneman and Tversky in 1979 (Kahneman & Tversky, 1979). No visible loss, no flinch.

Benartzi walks through the Save More Tomorrow design, flinch by flinch, in his TED talk.

Is saving really a discipline problem?

Mostly, no. In 2001, economists Brigitte Madrian and Dennis Shea studied a company that switched its 401(k) from opt-in to automatic enrollment. Same jobs, same employer match, one flipped default on a form. Participation among new hires jumped from 37% to 86% (Madrian & Shea, 2001).

Sit with that number. If saving were a character test, a form couldn’t move it 49 points. The workers didn’t change. The default did. The myth says undersavers lack discipline. The data says undersavers face a design that demands discipline, then blames them for not having it.

Here’s the part that rarely makes it into money-advice content: the retirement industry conceded this argument years ago. By 2024, 61% of Vanguard-administered plans enrolled workers automatically, up from about 10% in 2006 (Vanguard How America Saves, 2025). About two-thirds of automatic-enrollment plans now build in automatic annual increases (Vanguard, How America Saves 2025). The people managing trillions stopped betting on willpower two decades ago. Only the advice content still preaches it. the psychology behind your money habits

Thaler on nudges: why good design does the saving your willpower won't. Talks at Google.

Where does the trick fall short?

In two places, and both matter. A 2024 analysis by Choi, Laibson and colleagues followed savers across whole careers, not single years. Auto-enrollment’s steady-state effect came out at +0.6 percentage points of income, 72% smaller than the +2.2 points that year-one results implied (Choi, Laibson et al., 2024).

Why does the effect shrink? Life happens between paychecks. People change jobs, and every new employer resets the defaults. Many cash out small balances when they leave. Autopilot only flies while you’re on the plane, and most careers involve a lot of planes.

Autopilot, after the fine print Extra saving from automatic enrollment: year-one results extrapolated to 2.2 percentage points of income, but the steady-state effect across full careers was 0.6 percentage points, 72% smaller, per Choi, Laibson and colleagues, 2024. Autopilot, after the fine print Extra saving from auto-enrollment, percentage points of income Year-one extrapolation +2.2pp Career steady state +0.6pp (72% smaller) Source: Choi, Laibson et al. (2024), NBER Working Paper 32828
The honest math: still positive, much smaller than the demo. Source: Choi, Laibson et al. (2024).

The second catch hides inside the first study’s fine print. In Madrian and Shea’s data, 75% of auto-enrolled workers stayed parked at the 3% default rate (Madrian & Shea, 2001). The same inertia that pulls you in can pin you low. A default you never chose quietly becomes your ceiling.

So hold both truths at once. Design beats willpower: that part is settled. And design can anchor you to a number that’s too small: that part is settled too. Most content sells you the first half. The fix for the second half is exactly what SMarT added: automatic increases, plus one honest look at your rate each year.

How do you run Save More Tomorrow on yourself?

Copy the same three moves with any automatic transfer. The mechanism lives in the schedule, not in the account type. Momentum is on your side here: a record 45% of participants in Vanguard-administered plans raised their savings rate in 2024 (Vanguard How America Saves, 2025).

Heads up: this is psychology education, not financial advice. The moves below work on any auto-transfer: a savings account, an emergency fund, a retirement plan. What to put the money in is a question for a professional, not a blog.

Same three moves, translated:

  1. Commit before the trigger. Pick a future money event: your next raise, an annual review, even this coming January. Schedule a +1% increase to start on that date, today, while agreeing is cheap.
  2. Protect the paycheck rule. Route the increase out of new money before it ever reaches your checking account. If the number you live on never drops, the flinch never comes. route the raise before lifestyle creep eats it
  3. Keep the exit, add a floor. Give yourself standing permission to cancel. Then set one yearly reminder to check the rate, so a low default never becomes your ceiling.

A hand dropping a single coin into a blue piggy bank, one small automatic deposit at a time.

I run a bare-bones version of this with a plain savings account. One rule, written years ago: the month new money lands, a slice of it joins the automatic transfer before anything else gets planned. Most months I forget the rule exists. That’s the entire point. The system remembers so I don’t have to.

And if even the setup feels like a project, shrink it. Finding one setting in one app is a classic 2-minute action, the kind you size for your worst day, not your most organized one.

FAQ

Does this only work with a 401(k)?

No. The account is interchangeable; the design isn’t. Pre-commit to a future increase, tie it to new money, and keep the exit open. That combination moved savings rates from 3.5% to 13.6% in 40 months (Thaler & Benartzi, 2004), and it maps onto any automatic transfer you control.

What if my next raise is nowhere in sight?

Pick any future money event: a tax refund, an annual review, even January 1st. Present bias needs distance, not a raise. Committing months ahead is what got 78% of workers to say yes (Thaler & Benartzi, 2004). Schedule the increase now and let the calendar do the arguing.

Is auto-enrollment alone enough?

Usually not. In the classic study, 75% of auto-enrolled workers stayed parked at the 3% default (Madrian & Shea, 2001), and career-long effects shrink to about +0.6 percentage points of income (Choi, Laibson et al., 2024). Pair enrollment with automatic increases and one yearly rate check.

Did Save More Tomorrow participants actually stick with it?

Mostly, yes. 78% of workers offered the plan joined, and 80% of joiners stayed through four consecutive raises (Thaler & Benartzi, 2004). Even the people who quit kept their already-higher rate; they only stopped future increases. The design fails soft, which is part of the trick.

The Bottom Line

You’ve been told saving is a character test. The evidence says it’s a scheduling problem. That’s all Save More Tomorrow does: move the commitment into the future, and 78% say yes. Tie increases to raises and the sacrifice never registers as a loss. Flip a default and participation jumps from 37% to 86%.

Respect the fine print, though. Defaults anchor low, job changes leak savings, and the career-long gain is smaller than the demo. So don’t just switch on autopilot. Set the destination, then audit it once a year. Future-you isn’t a more disciplined person. Future-you is just you, minus the excuse.

Pre-commitment works because it removes the decision. Making future-you feel real is the other half, and the Future-You Savings Letter is that half on paper: a vividness minute, a letter in six parts, and one pre-decided action.

One small action today: open your retirement or savings app and find the automatic-increase setting. If it exists, schedule a +1% bump for your next raise date. If it doesn’t, set a calendar reminder for that date instead. Two minutes, then close the app.

Alex is the voice of Self Lab: practical psychology for people who are done with motivational fluff.

Sources