Short answer: the 4% rule comes from studies of 30-year retirements. In William Bengen’s 1994 study of US market history, a 4% first-year withdrawal, raised with inflation each year, never ran a 50/50 stock-and-bond portfolio dry in under 33 years. An early retirement can last 40 years or more, and research that looks at longer periods generally supports a lower starting rate. Morningstar’s 2025 model puts it at 3.9% for 30 years and 3.3% for 40 years. Planning on 3.5% instead of 4% means a target about 14% larger (28.6 times spending instead of 25), which in our example adds about a year and a half of saving. Neither rate guarantees the money lasts.
What the 4% rule actually says
It isn’t “withdraw 4% of your balance every year”. You take 4% of the portfolio’s value in the first year of retirement, then keep that dollar amount and raise it with inflation every year after, whatever the market does. On $1,000,000 that’s $40,000 in year one, about $41,200 in year two if inflation is 3%, and so on. Turned around, it gives the familiar FIRE number: spending divided by 4%, or 25 times what you spend in a year.
Where the 4% figure comes from
- Bengen, 1994. Using US stock and intermediate-term Treasury returns from 1926, Bengen tested inflation-adjusted withdrawals from a 50/50 portfolio. He found that a 4% first-year withdrawal had never exhausted a portfolio in under 33 years, that 3% to about 3.5% had always lasted at least 50 years, and that 4.25% could have run out in as little as 28 years (Journal of Financial Planning).
- The Trinity study, 1998. Three Trinity University professors measured how often portfolios survived payout periods of 15 to 30 years, using market returns from 1926 to 1995, and counted a period as a success if money remained at the end (AAII Journal). A later AAII review of the research summarized that a 4% inflation-adjusted withdrawal from a 50/50 portfolio “survived 95% of the 30-year periods” in that study (AAII Journal, 2008).
Both studies answer a 30-year question with historical US data, before fees and taxes. That is the setting the 4% rule was built for: a retirement starting in your early to mid 60s.
Why a longer retirement needs a lower rate
The longer the money has to last, the more bad stretches it has to survive. Morningstar’s 2025 research estimates the highest starting withdrawal rate that leaves money remaining in 90% of its simulated futures, using forward-looking return and inflation assumptions:
| Retirement length | 50% stocks | 100% stocks |
|---|---|---|
| 20 years | 5.3% | 4.6% |
| 30 years | 3.9% | 3.4% |
| 35 years | 3.5% | 3.2% |
| 40 years | 3.3% | 3.0% |
Two caveats keep this in proportion. Morningstar’s figures are model estimates built on its own forecasts, which it describes as conservative, and they move from year to year (its 30-year base case was 3.3% in 2021, 4.0% in 2023 and 3.9% in 2025). And “90% success” means the model still runs out of money in one future in ten. It’s a measure of risk, not a promise.
What choosing 3.5% costs you
A lower withdrawal rate means a bigger target, and a bigger target means more years of saving. Example: $80,000 take-home pay, $40,000 spending, $40,000 saved a year from a standing start, and a 5% return after inflation, the calculator’s default assumption, not a forecast.
| Withdrawal rate | Multiple of spending | FIRE number | Estimated years to reach it |
|---|---|---|---|
| 4.0% | 25.0× | $1,000,000 | 16.6 |
| 3.75% | 26.7× | $1,066,667 | 17.4 |
| 3.5% | 28.6× | $1,142,857 | 18.2 |
| 3.25% | 30.8× | $1,230,769 | 19.1 |
| 3.0% | 33.3× | $1,333,333 | 20.1 |
In this example, moving from 4% to 3.5% adds $142,857 to the target and about 1.6 years of work. The trade-off is that cost in exchange for more margin over a long retirement. You can run your own numbers by changing the withdrawal rate in the FIRE calculator.
Flexibility can matter more than the starting rate
The fixed-rate studies assume you never adjust, which few people do. Morningstar’s 2025 report finds that retirees who adopt a more flexible approach to spending could start as high as 5.7% of their starting balance, at the cost of income that rises and falls with the market. Other levers that make a given rate safer:
- spending you could cut in a bad year without hardship;
- some part-time or project income in the first years, when a market fall does the most damage (see sequence-of-returns risk);
- Social Security or a pension later in life, which reduces what the portfolio must cover;
- low fund fees, because the studies above ignore costs that come straight out of the return.
So which should you use?
It depends on how long the money must last and how much room you have to adjust. If your plan has little slack, no other income for decades and could run 40 years or more, the research supports planning below 4%, and 3.5% or lower is a reasonable starting point. If you can cut spending, earn a little, or expect a pension or Social Security to cover part of your costs later, a rate closer to 4% may be defensible. Either way, it’s a planning assumption to revisit every year, not a rule the market has to follow.
Sources
- William P. Bengen, “Determining Withdrawal Rates Using Historical Data,” Journal of Financial Planning, October 1994 (FPA reprint).
- Philip L. Cooley, Carl M. Hubbard and Daniel T. Walz, “Retirement Savings: Choosing a Withdrawal Rate That Is Sustainable,” AAII Journal, February 1998.
- William Reichenstein, “Will Your Savings Last? What the Withdrawal Rate Studies Show,” AAII Journal, July 2008.
- Amy C. Arnott and Christine Benz, The State of Retirement Income: 2025, Morningstar, December 3, 2025.
Related reading: Coast FIRE for a target you can hit sooner, and the Roth conversion ladder for drawing on retirement accounts before 59½.
