What is a safe withdrawal rate?
A safe withdrawal rate is a first-year withdrawal, as a share of your portfolio, that you then raise with inflation every year and that would have lasted your whole retirement in the historical periods you care about. William Bengen, who started this research in 1994, later coined "SAFEMAX" for the highest rate that lasted at least 30 years in every period in his data[1]. The well-known 4% rule is one safe withdrawal rate: the one for 30 years, a balanced US portfolio and a near-100% success target[2].
Change any of those conditions and the safe rate changes. This calculator asks for them directly, so instead of testing a rate you choose, it finds the rate that fits your situation.
How to use the safe withdrawal rate calculator
- Enter the age you retire and the age the money must last to, or switch to a number of years.
- Set your stock share and fees. The rest of the portfolio is 10-year US Treasuries, rebalanced every year.
- Pick a target success rate: the share of historical start years (and simulations) that must never run out. Add a final-balance goal if you want money left at the end.
- Read the safe rate, the table by horizon and mix, and the chart of every start year.
How does the calculator find your safe withdrawal rate?
For each retirement start year since 1928 with a complete history, it solves for the exact largest inflation-adjusted withdrawal that would have lasted your full horizon, using Damodaran's annual S&P 500, Treasury and inflation data[11]. Then it sorts those rates. With a 100% target, your safe rate is the lowest one, set by the worst start year. With a 95% target, the worst 5% of start years are allowed to fall short. The Monte Carlo test does the same with 2,000 retirements built from randomly drawn years.
A worked example with the defaults: retire at 55, plan to 95 (40 years), 80% stocks, 0.1% fees. The 59 start years could have paid between 3.53% (starting in 1966) and 9.87% (starting in 1982), with a median of 5.90%. Allowing the worst 2 to fall short gives 3.67%. Requiring every start year to last gives 3.53%.
Safe withdrawal rate by age
Your retirement age sets your horizon, and your horizon sets the rate. This table assumes you plan to age 95, with the calculator's defaults.
| Retire at | Years | Every start year | 95% of start years | Monte Carlo 95% |
|---|---|---|---|---|
| 35 | 60 | 3.30% | 3.44% | 2.39% |
| 40 | 55 | 3.33% | 3.47% | 2.45% |
| 45 | 50 | 3.37% | 3.50% | 2.63% |
| 50 | 45 | 3.44% | 3.57% | 2.66% |
| 55 | 40 | 3.53% | 3.67% | 2.85% |
| 60 | 35 | 3.64% | 3.79% | 3.04% |
| 65 | 30 | 3.77% | 3.96% | 3.30% |
| 70 | 25 | 4.07% | 4.28% | 3.85% |
Our calculation: 80% stocks, 0.1% fees, spending down to zero at 95, US data 1928-2025. Someone retiring at 65 had a 95% historical rate of 3.96%; someone retiring at 40 had 3.47%. For every one of these horizons, the worst start year was 1966.
Why do longer retirements need lower withdrawal rates?
Partly arithmetic. If your money earned nothing after inflation, you could withdraw 1/30 of it a year for 30 years (3.33%) but only 1/60 for 60 years (1.67%). At a steady 2% real return the gap narrows to 4.38% versus 2.82%, because returns pay for more of a long retirement. But returns are not steady, and that is the bigger reason.
A longer retirement has more room for a bad stretch, and a retirement that starts with one has to survive it for decades afterward. In our data, 4% with 80% stocks lasted in 94% of 30-year start years but only 82% of 60-year ones. Karsten Jeske found the same with data back to 1871: 4% with a 50/50 portfolio had about 95% success over 30 years and about 65% over 60[4]. He warns that 30-year results should not be stretched to longer retirements when the money may run down to zero[5], and concludes that rates of 3.25%-3.5% held up in the worst historical conditions over 50-60 years[7]. What matters most is the start: real returns in the first decade explain most of the difference between good and bad retirements[10].
How does your stock and bond mix change the safe rate?
Here is the historical safe rate at a 95% target for every combination of horizon and stock share, with Monte Carlo below each figure. The calculator recomputes it with your own target, fees and final-balance goal.
| Retirement | 40% stocks | 60% stocks | 80% stocks | 100% stocks |
|---|---|---|---|---|
| 30 years | 3.77%MC 3.49% | 3.85%MC 3.56% | 3.96%MC 3.30% | 3.90%MC 2.98% |
| 40 years | 3.39%MC 2.91% | 3.58%MC 3.00% | 3.67%MC 2.85% | 3.62%MC 2.46% |
| 50 years | 3.06%MC 2.60% | 3.40%MC 2.72% | 3.50%MC 2.63% | 3.47%MC 2.38% |
| 60 years | 2.93%MC 2.34% | 3.22%MC 2.46% | 3.44%MC 2.39% | 3.38%MC 2.13% |
In history, the mix with the highest safe rate was 80% stocks for every horizon in this table, and the 40% stock portfolio had the lowest rate at 60 years. That fits Bengen, who found allocations between 50% and 75% stocks best and called less than 50% counterproductive[2], and Jeske, who notes that bonds returned only about 2.6% a year after inflation since 1871, too little to carry a long retirement on their own[5]. More stocks also mean bigger swings in your balance along the way, so pick a mix you will actually hold through a crash.
What success rate should you target?
A stricter target protects you from worse start years but lowers what you can spend in every other one.
| Target (40 years) | Historical rate | First year on $1,000,000 | Monte Carlo rate |
|---|---|---|---|
| 80% | 4.22% | $42,200 | 4.11% |
| 90% | 3.92% | $39,200 | 3.35% |
| 95% | 3.67% | $36,700 | 2.85% |
| 100% | 3.53% | $35,300 | 2.04% (99%) |
Morningstar uses a 90% target for its 3.9% estimate[9], and the Trinity study reported success rates for every rate rather than picking one[3]. A 100% target is the only one that survived every start year on record, but it is sized for one worst case, and the future can be worse or better. A practical approach: choose 90%-95% if you could trim spending in a long downturn, and 100% if your spending is fixed. Monte Carlo can never certify 100%, because it can always draw a worse sequence than history, so its 100% column uses a 99% target.
How much does leaving money at the end cost?
The classic studies count a retirement as a success if a single dollar is left[3]. If you want to leave money to heirs, or keep a cushion in case you live longer than planned, set a final-balance goal.
| Left at the end | 30 years | 60 years |
|---|---|---|
| Nothing | 3.96% | 3.44% |
| 50% of the start | 3.55% | 3.37% |
| 100% of the start | 3.10% | 3.27% |
Historical rates at a 95% target, 80% stocks. Keeping the whole starting balance after inflation costs much more over 30 years than over 60, because over 60 years the portfolio has to stay large to keep paying anyway. Jeske found the same: with 100% stocks and 4% over 30 years, success fell from 97% to 80% once the full starting capital had to be preserved[5].
How do fees affect the safe withdrawal rate?
Every 1% in fees comes off every year's return. Over 40 years at a 95% target, the historical safe rate was 3.72% with no fees, 3.48% with 0.5% and 3.24% with 1%. Because fees work like a permanent cut in returns, the fee field doubles as a way to test lower future returns: raise it by the shortfall you expect.
Should you use a lower rate when stocks are expensive?
The historical safe rate is set by the worst past start years, and those tended to begin when stocks were expensive. Jeske found that every failure of the 4% rule over 30 years started when the Shiller CAPE ratio was above 15, and that 10-year real stock returns averaged below -1% after a CAPE above 30[6]. Morningstar bases its estimate on forward-looking return and inflation assumptions instead of history, which is why it moves from year to year: 3.3% in 2021, 4.0% in 2023 and 3.9% now[9].
This calculator does not adjust for valuations; it shows what history and a reshuffled history delivered. It also uses US data only, and US returns have been unusually kind to retirees: across 17 countries, a 4% real withdrawal held up for 30 years in only 4[1]. If either worries you, use the stricter target, the Monte Carlo rate or a higher fee setting.
Why is the Monte Carlo rate lower than the historical one?
History happened in one particular order, with only so many bad decades in it. The Monte Carlo test draws each year at random from the same 1928-2025 data, so it can string together several bad years that never happened in a row, and it ignores the order in which good and bad years actually followed each other. At low targets the two agree closely; at high targets and long horizons the Monte Carlo rate is lower because it finds sequences worse than any on record. Treat the historical rate as "what has worked" and the Monte Carlo rate as a stress test.
Safe withdrawal rate calculator vs the 4% rule calculator
They answer opposite questions. The 4% rule calculator starts from a rate, shows what it pays from your savings (or how much you need) and how often it lasted. This page starts from your situation and finds the rate. For the history and the research behind both, read our guide to the 4% rule, and for the rest of your plan, the FIRE calculator.